Search | arXiv e-print repository

<!DOCTYPE html> <html lang="en"> <head> <meta charset="utf-8"/> <meta name="viewport" content="width=device-width, initial-scale=1"/>  <link rel="apple-touch-icon" sizes="180x180" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/apple-touch-icon.png"> <link rel="icon" type="image/png" sizes="32x32" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon-32x32.png"> <link rel="icon" type="image/png" sizes="16x16" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon-16x16.png"> <link rel="manifest" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/site.webmanifest"> <link rel="mask-icon" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/safari-pinned-tab.svg" color="#b31b1b"> <link rel="shortcut icon" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon.ico"> <meta name="msapplication-TileColor" content="#b31b1b"> <meta name="msapplication-config" content="images/icons/browserconfig.xml"> <meta name="theme-color" content="#b31b1b">  <title>Search | arXiv e-print repository</title> <script defer src="https://static.arxiv.org/static/base/1.0.0a5/fontawesome-free-5.11.2-web/js/all.js"></script> <link rel="stylesheet" href="https://static.arxiv.org/static/base/1.0.0a5/css/arxivstyle.css" /> <script type="text/x-mathjax-config"> MathJax.Hub.Config({ messageStyle: "none", extensions: ["tex2jax.js"], jax: ["input/TeX", "output/HTML-CSS"], tex2jax: { inlineMath: [ ['$','$'], ["\$","\$"] ], displayMath: [ ['$$','$$'], ["\\[","\\]"] ], processEscapes: true, ignoreClass: '.*', processClass: 'mathjax.*' }, TeX: { extensions: ["AMSmath.js", "AMSsymbols.js", "noErrors.js"], noErrors: { inlineDelimiters: ["$","$"], multiLine: false, style: { "font-size": "normal", "border": "" } } }, "HTML-CSS": { availableFonts: ["TeX"] } }); </script> <script src='//static.arxiv.org/MathJax-2.7.3/MathJax.js'></script> <script src="https://static.arxiv.org/static/base/1.0.0a5/js/notification.js"></script> <link rel="stylesheet" href="https://static.arxiv.org/static/search/0.5.6/css/bulma-tooltip.min.css" /> <link rel="stylesheet" href="https://static.arxiv.org/static/search/0.5.6/css/search.css" /> <script src="https://code.jquery.com/jquery-3.2.1.slim.min.js" integrity="sha256-k2WSCIexGzOj3Euiig+TlR8gA0EmPjuc79OEeY5L45g=" crossorigin="anonymous"></script> <script src="https://static.arxiv.org/static/search/0.5.6/js/fieldset.js"></script> <style> radio#cf-customfield_11400 { display: none; } </style> </head> <body> <header><a href="#main-container" class="is-sr-only">Skip to main content</a>  <div class="attribution level is-marginless" role="banner"> <div class="level-left"> <a class="level-item" href="https://cornell.edu/"><img src="https://static.arxiv.org/static/base/1.0.0a5/images/cornell-reduced-white-SMALL.svg" alt="Cornell University" width="200" aria-label="logo" /></a> </div> <div class="level-right is-marginless"><p class="sponsors level-item is-marginless"><span id="support-ack-url">We gratefully acknowledge support from<br /> the Simons Foundation, <a href="https://info.arxiv.org/about/ourmembers.html">member institutions</a>, and all contributors. <a href="https://info.arxiv.org/about/donate.html">Donate</a></span></p></div> </div>  <div class="identity level is-marginless"> <div class="level-left"> <div class="level-item"> <a class="arxiv" href="https://arxiv.org/" aria-label="arxiv-logo"> <img src="https://static.arxiv.org/static/base/1.0.0a5/images/arxiv-logo-one-color-white.svg" aria-label="logo" alt="arxiv logo" width="85" style="width:85px;"/> </a> </div> </div> <div class="search-block level-right"> <form class="level-item mini-search" method="GET" action="https://arxiv.org/search"> <div class="field has-addons"> <div class="control"> <input class="input is-small" type="text" name="query" placeholder="Search..." aria-label="Search term or terms" /> <p class="help"><a href="https://info.arxiv.org/help">Help</a> | <a href="https://arxiv.org/search/advanced">Advanced Search</a></p> </div> <div class="control"> <div class="select is-small"> <select name="searchtype" aria-label="Field to search"> <option value="all" selected="selected">All fields</option> <option value="title">Title</option> <option value="author">Author</option> <option value="abstract">Abstract</option> <option value="comments">Comments</option> <option value="journal_ref">Journal reference</option> <option value="acm_class">ACM classification</option> <option value="msc_class">MSC classification</option> <option value="report_num">Report number</option> <option value="paper_id">arXiv identifier</option> <option value="doi">DOI</option> <option value="orcid">ORCID</option> <option value="author_id">arXiv author ID</option> <option value="help">Help pages</option> <option value="full_text">Full text</option> </select> </div> </div> <input type="hidden" name="source" value="header"> <button class="button is-small is-cul-darker">Search</button> </div> </form> </div> </div>  <div class="container"> <div class="user-tools is-size-7 has-text-right has-text-weight-bold" role="navigation" aria-label="User menu"> <a href="https://arxiv.org/login">Login</a> </div> </div> </header> <main class="container" id="main-container"> <div class="level is-marginless"> <div class="level-left"> <h1 class="title is-clearfix"> Showing 1–50 of 2,203 results for author: <span class="mathjax">Yang, X</span> </h1> </div> <div class="level-right is-hidden-mobile">  <span class="help" style="display: inline-block;"><a href="https://github.com/arXiv/arxiv-search/releases">Search v0.5.6 released 2020-02-24</a>  </span> </div> </div> <div class="content"> <form method="GET" action="/search/cs" aria-role="search"> Searching in archive <strong>cs</strong>. <a href="/search/?searchtype=author&query=Yang%2C+X">Search in all archives.</a> <div class="field has-addons-tablet"> <div class="control is-expanded"> <label for="query" class="hidden-label">Search term or terms</label> <input class="input is-medium" id="query" name="query" placeholder="Search term..." type="text" value="Yang, X"> </div> <div class="select control is-medium"> <label class="is-hidden" for="searchtype">Field</label> <select class="is-medium" id="searchtype" name="searchtype"><option value="all">All fields</option><option value="title">Title</option><option selected value="author">Author(s)</option><option value="abstract">Abstract</option><option value="comments">Comments</option><option value="journal_ref">Journal reference</option><option value="acm_class">ACM classification</option><option value="msc_class">MSC classification</option><option value="report_num">Report number</option><option value="paper_id">arXiv identifier</option><option value="doi">DOI</option><option value="orcid">ORCID</option><option value="license">License (URI)</option><option value="author_id">arXiv author ID</option><option value="help">Help pages</option><option value="full_text">Full text</option></select> </div> <div class="control"> <button class="button is-link is-medium">Search</button> </div> </div> <div class="field"> <div class="control is-size-7"> <label class="radio"> <input checked id="abstracts-0" name="abstracts" type="radio" value="show"> Show abstracts </label> <label class="radio"> <input id="abstracts-1" name="abstracts" type="radio" value="hide"> Hide abstracts </label> </div> </div> <div class="is-clearfix" style="height: 2.5em"> <div class="is-pulled-right"> <a href="/search/advanced?terms-0-term=Yang%2C+X&terms-0-field=author&size=50&order=-announced_date_first">Advanced Search</a> </div> </div> <input type="hidden" name="order" value="-announced_date_first"> <input type="hidden" name="size" value="50"> </form> <div class="level breathe-horizontal"> <div class="level-left"> <form method="GET" action="/search/"> <div style="display: none;"> <select id="searchtype" name="searchtype"><option value="all">All fields</option><option value="title">Title</option><option selected value="author">Author(s)</option><option value="abstract">Abstract</option><option value="comments">Comments</option><option value="journal_ref">Journal reference</option><option value="acm_class">ACM classification</option><option value="msc_class">MSC classification</option><option value="report_num">Report number</option><option value="paper_id">arXiv identifier</option><option value="doi">DOI</option><option value="orcid">ORCID</option><option value="license">License (URI)</option><option value="author_id">arXiv author ID</option><option value="help">Help pages</option><option value="full_text">Full text</option></select> <input id="query" name="query" type="text" value="Yang, X"> <ul id="abstracts"><li><input checked id="abstracts-0" name="abstracts" type="radio" value="show"> <label for="abstracts-0">Show abstracts</label></li><li><input id="abstracts-1" name="abstracts" type="radio" value="hide"> <label for="abstracts-1">Hide abstracts</label></li></ul> </div> <div class="box field is-grouped is-grouped-multiline level-item"> <div class="control"> <span class="select is-small"> <select id="size" name="size"><option value="25">25</option><option selected value="50">50</option><option value="100">100</option><option value="200">200</option></select> </span> <label for="size">results per page</label>. </div> <div class="control"> <label for="order">Sort results by</label> <span class="select is-small"> <select id="order" name="order"><option selected value="-announced_date_first">Announcement date (newest first)</option><option value="announced_date_first">Announcement date (oldest first)</option><option value="-submitted_date">Submission date (newest first)</option><option value="submitted_date">Submission date (oldest first)</option><option value="">Relevance</option></select> </span> </div> <div class="control"> <button class="button is-small is-link">Go</button> </div> </div> </form> </div> </div> <nav class="pagination is-small is-centered breathe-horizontal" role="navigation" aria-label="pagination"> <a href="" class="pagination-previous is-invisible">Previous </a> <a href="/search/?searchtype=author&query=Yang%2C+X&start=50" class="pagination-next" >Next </a> <ul class="pagination-list"> <li> <a href="/search/?searchtype=author&query=Yang%2C+X&start=0" class="pagination-link is-current" aria-label="Goto page 1">1 </a> </li> <li> <a href="/search/?searchtype=author&query=Yang%2C+X&start=50" class="pagination-link " aria-label="Page 2" aria-current="page">2 </a> </li> <li> <a href="/search/?searchtype=author&query=Yang%2C+X&start=100" class="pagination-link " aria-label="Page 3" aria-current="page">3 </a> </li> <li> <a href="/search/?searchtype=author&query=Yang%2C+X&start=150" class="pagination-link " aria-label="Page 4" aria-current="page">4 </a> </li> <li> <a href="/search/?searchtype=author&query=Yang%2C+X&start=200" class="pagination-link " aria-label="Page 5" aria-current="page">5 </a> </li> <li><span class="pagination-ellipsis">…</span></li> </ul> </nav> <ol class="breathe-horizontal" start="1"> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.14401">arXiv:2411.14401</a> <span> [<a href="https://arxiv.org/pdf/2411.14401">pdf</a>, <a href="https://arxiv.org/format/2411.14401">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Beyond Training: Dynamic Token Merging for Zero-Shot Video Understanding </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhang%2C+Y">Yiming Zhang</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+Z">Zhuokai Zhao</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Z">Zhaorun Chen</a>, <a href="/search/cs?searchtype=author&query=Ding%2C+Z">Zenghui Ding</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xianjun Yang</a>, <a href="/search/cs?searchtype=author&query=Sun%2C+Y">Yining Sun</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.14401v1-abstract-short" style="display: inline;"> Recent advancements in multimodal large language models (MLLMs) have opened new avenues for video understanding. However, achieving high fidelity in zero-shot video tasks remains challenging. Traditional video processing methods rely heavily on fine-tuning to capture nuanced spatial-temporal details, which incurs significant data and computation costs. In contrast, training-free approaches, though… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.14401v1-abstract-full').style.display = 'inline'; document.getElementById('2411.14401v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.14401v1-abstract-full" style="display: none;"> Recent advancements in multimodal large language models (MLLMs) have opened new avenues for video understanding. However, achieving high fidelity in zero-shot video tasks remains challenging. Traditional video processing methods rely heavily on fine-tuning to capture nuanced spatial-temporal details, which incurs significant data and computation costs. In contrast, training-free approaches, though efficient, often lack robustness in preserving context-rich features across complex video content. To this end, we propose DYTO, a novel dynamic token merging framework for zero-shot video understanding that adaptively optimizes token efficiency while preserving crucial scene details. DYTO integrates a hierarchical frame selection and a bipartite token merging strategy to dynamically cluster key frames and selectively compress token sequences, striking a balance between computational efficiency with semantic richness. Extensive experiments across multiple benchmarks demonstrate the effectiveness of DYTO, achieving superior performance compared to both fine-tuned and training-free methods and setting a new state-of-the-art for zero-shot video understanding. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.14401v1-abstract-full').style.display = 'none'; document.getElementById('2411.14401v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 21 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.13577">arXiv:2411.13577</a> <span> [<a href="https://arxiv.org/pdf/2411.13577">pdf</a>, <a href="https://arxiv.org/format/2411.13577">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Audio and Speech Processing">eess.AS</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Multimedia">cs.MM</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Sound">cs.SD</span> </div> </div> <p class="title is-5 mathjax"> WavChat: A Survey of Spoken Dialogue Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ji%2C+S">Shengpeng Ji</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Y">Yifu Chen</a>, <a href="/search/cs?searchtype=author&query=Fang%2C+M">Minghui Fang</a>, <a href="/search/cs?searchtype=author&query=Zuo%2C+J">Jialong Zuo</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+J">Jingyu Lu</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+H">Hanting Wang</a>, <a href="/search/cs?searchtype=author&query=Jiang%2C+Z">Ziyue Jiang</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+L">Long Zhou</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+S">Shujie Liu</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+X">Xize Cheng</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaoda Yang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Z">Zehan Wang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+Q">Qian Yang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+J">Jian Li</a>, <a href="/search/cs?searchtype=author&query=Jiang%2C+Y">Yidi Jiang</a>, <a href="/search/cs?searchtype=author&query=He%2C+J">Jingzhen He</a>, <a href="/search/cs?searchtype=author&query=Chu%2C+Y">Yunfei Chu</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+J">Jin Xu</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+Z">Zhou Zhao</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.13577v1-abstract-short" style="display: inline;"> Recent advancements in spoken dialogue models, exemplified by systems like GPT-4o, have captured significant attention in the speech domain. Compared to traditional three-tier cascaded spoken dialogue models that comprise speech recognition (ASR), large language models (LLMs), and text-to-speech (TTS), modern spoken dialogue models exhibit greater intelligence. These advanced spoken dialogue model… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.13577v1-abstract-full').style.display = 'inline'; document.getElementById('2411.13577v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.13577v1-abstract-full" style="display: none;"> Recent advancements in spoken dialogue models, exemplified by systems like GPT-4o, have captured significant attention in the speech domain. Compared to traditional three-tier cascaded spoken dialogue models that comprise speech recognition (ASR), large language models (LLMs), and text-to-speech (TTS), modern spoken dialogue models exhibit greater intelligence. These advanced spoken dialogue models not only comprehend audio, music, and other speech-related features, but also capture stylistic and timbral characteristics in speech. Moreover, they generate high-quality, multi-turn speech responses with low latency, enabling real-time interaction through simultaneous listening and speaking capability. Despite the progress in spoken dialogue systems, there is a lack of comprehensive surveys that systematically organize and analyze these systems and the underlying technologies. To address this, we have first compiled existing spoken dialogue systems in the chronological order and categorized them into the cascaded and end-to-end paradigms. We then provide an in-depth overview of the core technologies in spoken dialogue models, covering aspects such as speech representation, training paradigm, streaming, duplex, and interaction capabilities. Each section discusses the limitations of these technologies and outlines considerations for future research. Additionally, we present a thorough review of relevant datasets, evaluation metrics, and benchmarks from the perspectives of training and evaluating spoken dialogue systems. We hope this survey will contribute to advancing both academic research and industrial applications in the field of spoken dialogue systems. The related material is available at https://github.com/jishengpeng/WavChat. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.13577v1-abstract-full').style.display = 'none'; document.getElementById('2411.13577v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 14 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">60 papes, working in progress</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.13116">arXiv:2411.13116</a> <span> [<a href="https://arxiv.org/pdf/2411.13116">pdf</a>, <a href="https://arxiv.org/format/2411.13116">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Provably Efficient Action-Manipulation Attack Against Continuous Reinforcement Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Luo%2C+Z">Zhi Luo</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiyuan Yang</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+P">Pan Zhou</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+D">Di Wang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.13116v1-abstract-short" style="display: inline;"> Manipulating the interaction trajectories between the intelligent agent and the environment can control the agent's training and behavior, exposing the potential vulnerabilities of reinforcement learning (RL). For example, in Cyber-Physical Systems (CPS) controlled by RL, the attacker can manipulate the actions of the adopted RL to other actions during the training phase, which will lead to bad co… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.13116v1-abstract-full').style.display = 'inline'; document.getElementById('2411.13116v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.13116v1-abstract-full" style="display: none;"> Manipulating the interaction trajectories between the intelligent agent and the environment can control the agent's training and behavior, exposing the potential vulnerabilities of reinforcement learning (RL). For example, in Cyber-Physical Systems (CPS) controlled by RL, the attacker can manipulate the actions of the adopted RL to other actions during the training phase, which will lead to bad consequences. Existing work has studied action-manipulation attacks in tabular settings, where the states and actions are discrete. As seen in many up-and-coming RL applications, such as autonomous driving, continuous action space is widely accepted, however, its action-manipulation attacks have not been thoroughly investigated yet. In this paper, we consider this crucial problem in both white-box and black-box scenarios. Specifically, utilizing the knowledge derived exclusively from trajectories, we propose a black-box attack algorithm named LCBT, which uses the Monte Carlo tree search method for efficient action searching and manipulation. Additionally, we demonstrate that for an agent whose dynamic regret is sub-linearly related to the total number of steps, LCBT can teach the agent to converge to target policies with only sublinear attack cost, i.e., $O\left(\mathcal{R}(T) + MH^3K^E\log (MT)\right)(0<E<1)$, where $H$ is the number of steps per episode, $K$ is the total number of episodes, $T=KH$ is the total number of steps, $M$ is the number of subspaces divided in the state space, and $\mathcal{R}(T)$ is the bound of the RL algorithm's regret. We conduct our proposed attack methods on three aggressive algorithms: DDPG, PPO, and TD3 in continuous settings, which show a promising attack performance. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.13116v1-abstract-full').style.display = 'none'; document.getElementById('2411.13116v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 20 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.13093">arXiv:2411.13093</a> <span> [<a href="https://arxiv.org/pdf/2411.13093">pdf</a>, <a href="https://arxiv.org/format/2411.13093">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Video-RAG: Visually-aligned Retrieval-Augmented Long Video Comprehension </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Luo%2C+Y">Yongdong Luo</a>, <a href="/search/cs?searchtype=author&query=Zheng%2C+X">Xiawu Zheng</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiao Yang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+G">Guilin Li</a>, <a href="/search/cs?searchtype=author&query=Lin%2C+H">Haojia Lin</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+J">Jinfa Huang</a>, <a href="/search/cs?searchtype=author&query=Ji%2C+J">Jiayi Ji</a>, <a href="/search/cs?searchtype=author&query=Chao%2C+F">Fei Chao</a>, <a href="/search/cs?searchtype=author&query=Luo%2C+J">Jiebo Luo</a>, <a href="/search/cs?searchtype=author&query=Ji%2C+R">Rongrong Ji</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.13093v1-abstract-short" style="display: inline;"> Existing large video-language models (LVLMs) struggle to comprehend long videos correctly due to limited context. To address this problem, fine-tuning long-context LVLMs and employing GPT-based agents have emerged as promising solutions. However, fine-tuning LVLMs would require extensive high-quality data and substantial GPU resources, while GPT-based agents would rely on proprietary models (e.g.,… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.13093v1-abstract-full').style.display = 'inline'; document.getElementById('2411.13093v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.13093v1-abstract-full" style="display: none;"> Existing large video-language models (LVLMs) struggle to comprehend long videos correctly due to limited context. To address this problem, fine-tuning long-context LVLMs and employing GPT-based agents have emerged as promising solutions. However, fine-tuning LVLMs would require extensive high-quality data and substantial GPU resources, while GPT-based agents would rely on proprietary models (e.g., GPT-4o). In this paper, we propose Video Retrieval-Augmented Generation (Video-RAG), a training-free and cost-effective pipeline that employs visually-aligned auxiliary texts to help facilitate cross-modality alignment while providing additional information beyond the visual content. Specifically, we leverage open-source external tools to extract visually-aligned information from pure video data (e.g., audio, optical character, and object detection), and incorporate the extracted information into an existing LVLM as auxiliary texts, alongside video frames and queries, in a plug-and-play manner. Our Video-RAG offers several key advantages: (i) lightweight with low computing overhead due to single-turn retrieval; (ii) easy implementation and compatibility with any LVLM; and (iii) significant, consistent performance gains across long video understanding benchmarks, including Video-MME, MLVU, and LongVideoBench. Notably, our model demonstrates superior performance over proprietary models like Gemini-1.5-Pro and GPT-4o when utilized with a 72B model. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.13093v1-abstract-full').style.display = 'none'; document.getElementById('2411.13093v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 20 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">10 pages, 6 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.13050">arXiv:2411.13050</a> <span> [<a href="https://arxiv.org/pdf/2411.13050">pdf</a>, <a href="https://arxiv.org/format/2411.13050">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Hardware Architecture">cs.AR</span> </div> </div> <p class="title is-5 mathjax"> Topkima-Former: Low-energy, Low-Latency Inference for Transformers using top-k In-memory ADC </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Dong%2C+S">Shuai Dong</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+J">Junyi Yang</a>, <a href="/search/cs?searchtype=author&query=Peng%2C+X">Xiaoqi Peng</a>, <a href="/search/cs?searchtype=author&query=Shang%2C+H">Hongyang Shang</a>, <a href="/search/cs?searchtype=author&query=Ke%2C+Y">Ye Ke</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaofeng Yang</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+H">Hongjie Liu</a>, <a href="/search/cs?searchtype=author&query=Basu%2C+A">Arindam Basu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.13050v1-abstract-short" style="display: inline;"> Transformer model has gained prominence as a popular deep neural network architecture for neural language processing (NLP) and computer vision (CV) applications. However, the extensive use of nonlinear operations, like softmax, poses a performance bottleneck during transformer inference and comprises up to 40% of the total latency. Hence, we propose innovations at the circuit, architecture, and al… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.13050v1-abstract-full').style.display = 'inline'; document.getElementById('2411.13050v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.13050v1-abstract-full" style="display: none;"> Transformer model has gained prominence as a popular deep neural network architecture for neural language processing (NLP) and computer vision (CV) applications. However, the extensive use of nonlinear operations, like softmax, poses a performance bottleneck during transformer inference and comprises up to 40% of the total latency. Hence, we propose innovations at the circuit, architecture, and algorithm levels to accelerate the transformer. At the circuit level, we propose topkima-combining top-k activation selection with in-memory ADC (IMA) to implement a low-energy and low-latency softmax without any sorting latency. Only the k largest activations are sent to the softmax calculation block, reducing the huge computational cost of softmax. Using a modified training scheme with top-k only in the forward pass, experimental results demonstrate only a 0.4% to 1.2% reduction in accuracy across ViT, distilBERT, and BERT-base models when evaluated on CIFAR-10, CIFAR-100, and SQuAD datasets with k=5. At the architecture level, an improved scale-free technique is introduced to reduce the computational cost of attention. The combined system, dubbed Topkima-Former, enhances 1.8x-84x speedup and 1.3x-35x energy efficiency (EE) over prior In-memory computing (IMC) accelerators. Compared to a conventional softmax macro and a digital top-k (Dtopk) softmax macro, our proposed tokima softmax macro achieves about 15x and 8x faster speed respectively. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.13050v1-abstract-full').style.display = 'none'; document.getElementById('2411.13050v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 20 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">7 pages</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.12790">arXiv:2411.12790</a> <span> [<a href="https://arxiv.org/pdf/2411.12790">pdf</a>, <a href="https://arxiv.org/format/2411.12790">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Visual-Oriented Fine-Grained Knowledge Editing for MultiModal Large Language Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zeng%2C+Z">Zhen Zeng</a>, <a href="/search/cs?searchtype=author&query=Gu%2C+L">Leijiang Gu</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xun Yang</a>, <a href="/search/cs?searchtype=author&query=Duan%2C+Z">Zhangling Duan</a>, <a href="/search/cs?searchtype=author&query=Shi%2C+Z">Zenglin Shi</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+M">Meng Wang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.12790v1-abstract-short" style="display: inline;"> Knowledge editing aims to efficiently and cost-effectively correct inaccuracies and update outdated information. Recently, there has been growing interest in extending knowledge editing from Large Language Models (LLMs) to Multimodal Large Language Models (MLLMs), which integrate both textual and visual information, introducing additional editing complexities. Existing multimodal knowledge editing… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.12790v1-abstract-full').style.display = 'inline'; document.getElementById('2411.12790v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.12790v1-abstract-full" style="display: none;"> Knowledge editing aims to efficiently and cost-effectively correct inaccuracies and update outdated information. Recently, there has been growing interest in extending knowledge editing from Large Language Models (LLMs) to Multimodal Large Language Models (MLLMs), which integrate both textual and visual information, introducing additional editing complexities. Existing multimodal knowledge editing works primarily focus on text-oriented, coarse-grained scenarios, failing to address the unique challenges posed by multimodal contexts. In this paper, we propose a visual-oriented, fine-grained multimodal knowledge editing task that targets precise editing in images with multiple interacting entities. We introduce the Fine-Grained Visual Knowledge Editing (FGVEdit) benchmark to evaluate this task. Moreover, we propose a Multimodal Scope Classifier-based Knowledge Editor (MSCKE) framework. MSCKE leverages a multimodal scope classifier that integrates both visual and textual information to accurately identify and update knowledge related to specific entities within images. This approach ensures precise editing while preserving irrelevant information, overcoming the limitations of traditional text-only editing methods. Extensive experiments on the FGVEdit benchmark demonstrate that MSCKE outperforms existing methods, showcasing its effectiveness in solving the complex challenges of multimodal knowledge editing. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.12790v1-abstract-full').style.display = 'none'; document.getElementById('2411.12790v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 19 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.11941">arXiv:2411.11941</a> <span> [<a href="https://arxiv.org/pdf/2411.11941">pdf</a>, <a href="https://arxiv.org/format/2411.11941">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> TimeFormer: Capturing Temporal Relationships of Deformable 3D Gaussians for Robust Reconstruction </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jiang%2C+D">DaDong Jiang</a>, <a href="/search/cs?searchtype=author&query=Ke%2C+Z">Zhihui Ke</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+X">Xiaobo Zhou</a>, <a href="/search/cs?searchtype=author&query=Hou%2C+Z">Zhi Hou</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xianghui Yang</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+W">Wenbo Hu</a>, <a href="/search/cs?searchtype=author&query=Qiu%2C+T">Tie Qiu</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+C">Chunchao Guo</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.11941v1-abstract-short" style="display: inline;"> Dynamic scene reconstruction is a long-term challenge in 3D vision. Recent methods extend 3D Gaussian Splatting to dynamic scenes via additional deformation fields and apply explicit constraints like motion flow to guide the deformation. However, they learn motion changes from individual timestamps independently, making it challenging to reconstruct complex scenes, particularly when dealing with v… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11941v1-abstract-full').style.display = 'inline'; document.getElementById('2411.11941v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.11941v1-abstract-full" style="display: none;"> Dynamic scene reconstruction is a long-term challenge in 3D vision. Recent methods extend 3D Gaussian Splatting to dynamic scenes via additional deformation fields and apply explicit constraints like motion flow to guide the deformation. However, they learn motion changes from individual timestamps independently, making it challenging to reconstruct complex scenes, particularly when dealing with violent movement, extreme-shaped geometries, or reflective surfaces. To address the above issue, we design a plug-and-play module called TimeFormer to enable existing deformable 3D Gaussians reconstruction methods with the ability to implicitly model motion patterns from a learning perspective. Specifically, TimeFormer includes a Cross-Temporal Transformer Encoder, which adaptively learns the temporal relationships of deformable 3D Gaussians. Furthermore, we propose a two-stream optimization strategy that transfers the motion knowledge learned from TimeFormer to the base stream during the training phase. This allows us to remove TimeFormer during inference, thereby preserving the original rendering speed. Extensive experiments in the multi-view and monocular dynamic scenes validate qualitative and quantitative improvement brought by TimeFormer. Project Page: https://patrickddj.github.io/TimeFormer/ <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11941v1-abstract-full').style.display = 'none'; document.getElementById('2411.11941v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.11551">arXiv:2411.11551</a> <span> [<a href="https://arxiv.org/pdf/2411.11551">pdf</a>, <a href="https://arxiv.org/format/2411.11551">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Cryptography and Security">cs.CR</span> </div> </div> <p class="title is-5 mathjax"> Simple But Not Secure: An Empirical Security Analysis of Two-factor Authentication Systems </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wang%2C+Z">Zhi Wang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xin Yang</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+D">Du Chen</a>, <a href="/search/cs?searchtype=author&query=Gao%2C+H">Han Gao</a>, <a href="/search/cs?searchtype=author&query=Tian%2C+M">Meiqi Tian</a>, <a href="/search/cs?searchtype=author&query=Jia%2C+Y">Yan Jia</a>, <a href="/search/cs?searchtype=author&query=Li%2C+W">Wanpeng Li</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.11551v1-abstract-short" style="display: inline;"> To protect users from data breaches and phishing attacks, service providers typically implement two-factor authentication (2FA) to add an extra layer of security against suspicious login attempts. However, since 2FA can sometimes hinder user experience by introducing additional steps, many websites aim to reduce inconvenience by minimizing the frequency of 2FA prompts. One approach to achieve this… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11551v1-abstract-full').style.display = 'inline'; document.getElementById('2411.11551v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.11551v1-abstract-full" style="display: none;"> To protect users from data breaches and phishing attacks, service providers typically implement two-factor authentication (2FA) to add an extra layer of security against suspicious login attempts. However, since 2FA can sometimes hinder user experience by introducing additional steps, many websites aim to reduce inconvenience by minimizing the frequency of 2FA prompts. One approach to achieve this is by storing the user's ``Remember the Device'' preference in a cookie. As a result, users are only prompted for 2FA when this cookie expires or if they log in from a new device. To understand and improve the security of 2FA systems in real-world settings, we propose SE2FA, a vulnerability evaluation framework designed to detect vulnerabilities in 2FA systems. This framework enables us to analyze the security of 407 2FA systems across popular websites from the Tranco Top 10,000 list. Our analysis and evaluation found three zero-day vulnerabilities on three service providers that could allow an attacker to access a victim's account without possessing the victim's second authentication factor, thereby bypassing 2FA protections entirely. A further investigation found that these vulnerabilities stem from design choices aimed at simplifying 2FA for users but that unintentionally reduce its security effectiveness. We have disclosed these findings to the affected websites and assisted them in mitigating the risks. Based on the insights from this research, we provide practical recommendations for countermeasures to strengthen 2FA security and address these newly identified threats. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11551v1-abstract-full').style.display = 'none'; document.getElementById('2411.11551v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.11407">arXiv:2411.11407</a> <span> [<a href="https://arxiv.org/pdf/2411.11407">pdf</a>, <a href="https://arxiv.org/format/2411.11407">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> The Dark Side of Trust: Authority Citation-Driven Jailbreak Attacks on Large Language Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xikang Yang</a>, <a href="/search/cs?searchtype=author&query=Tang%2C+X">Xuehai Tang</a>, <a href="/search/cs?searchtype=author&query=Han%2C+J">Jizhong Han</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+S">Songlin Hu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.11407v1-abstract-short" style="display: inline;"> The widespread deployment of large language models (LLMs) across various domains has showcased their immense potential while exposing significant safety vulnerabilities. A major concern is ensuring that LLM-generated content aligns with human values. Existing jailbreak techniques reveal how this alignment can be compromised through specific prompts or adversarial suffixes. In this study, we introd… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11407v1-abstract-full').style.display = 'inline'; document.getElementById('2411.11407v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.11407v1-abstract-full" style="display: none;"> The widespread deployment of large language models (LLMs) across various domains has showcased their immense potential while exposing significant safety vulnerabilities. A major concern is ensuring that LLM-generated content aligns with human values. Existing jailbreak techniques reveal how this alignment can be compromised through specific prompts or adversarial suffixes. In this study, we introduce a new threat: LLMs' bias toward authority. While this inherent bias can improve the quality of outputs generated by LLMs, it also introduces a potential vulnerability, increasing the risk of producing harmful content. Notably, the biases in LLMs is the varying levels of trust given to different types of authoritative information in harmful queries. For example, malware development often favors trust GitHub. To better reveal the risks with LLM, we propose DarkCite, an adaptive authority citation matcher and generator designed for a black-box setting. DarkCite matches optimal citation types to specific risk types and generates authoritative citations relevant to harmful instructions, enabling more effective jailbreak attacks on aligned LLMs.Our experiments show that DarkCite achieves a higher attack success rate (e.g., LLama-2 at 76% versus 68%) than previous methods. To counter this risk, we propose an authenticity and harm verification defense strategy, raising the average defense pass rate (DPR) from 11% to 74%. More importantly, the ability to link citations to the content they encompass has become a foundational function in LLMs, amplifying the influence of LLMs' bias toward authority. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11407v1-abstract-full').style.display = 'none'; document.getElementById('2411.11407v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.11343">arXiv:2411.11343</a> <span> [<a href="https://arxiv.org/pdf/2411.11343">pdf</a>, <a href="https://arxiv.org/format/2411.11343">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Applications">stat.AP</span> </div> </div> <p class="title is-5 mathjax"> Teaching Video Diffusion Model with Latent Physical Phenomenon Knowledge </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Cao%2C+Q">Qinglong Cao</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+D">Ding Wang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+X">Xirui Li</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Y">Yuntian Chen</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+C">Chao Ma</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaokang Yang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.11343v1-abstract-short" style="display: inline;"> Video diffusion models have exhibited tremendous progress in various video generation tasks. However, existing models struggle to capture latent physical knowledge, failing to infer physical phenomena that are challenging to articulate with natural language. Generating videos following the fundamental physical laws is still an opening challenge. To address this challenge, we propose a novel method… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11343v1-abstract-full').style.display = 'inline'; document.getElementById('2411.11343v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.11343v1-abstract-full" style="display: none;"> Video diffusion models have exhibited tremendous progress in various video generation tasks. However, existing models struggle to capture latent physical knowledge, failing to infer physical phenomena that are challenging to articulate with natural language. Generating videos following the fundamental physical laws is still an opening challenge. To address this challenge, we propose a novel method to teach video diffusion models with latent physical phenomenon knowledge, enabling the accurate generation of physically informed phenomena. Specifically, we first pretrain Masked Autoencoders (MAE) to reconstruct the physical phenomena, resulting in output embeddings that encapsulate latent physical phenomenon knowledge. Leveraging these embeddings, we could generate the pseudo-language prompt features based on the aligned spatial relationships between CLIP vision and language encoders. Particularly, given that diffusion models typically use CLIP's language encoder for text prompt embeddings, our approach integrates the CLIP visual features informed by latent physical knowledge into a quaternion hidden space. This enables the modeling of spatial relationships to produce physical knowledge-informed pseudo-language prompts. By incorporating these prompt features and fine-tuning the video diffusion model in a parameter-efficient manner, the physical knowledge-informed videos are successfully generated. We validate our method extensively through both numerical simulations and real-world observations of physical phenomena, demonstrating its remarkable performance across diverse scenarios. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11343v1-abstract-full').style.display = 'none'; document.getElementById('2411.11343v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">7 figures, 14 pages</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.11318">arXiv:2411.11318</a> <span> [<a href="https://arxiv.org/pdf/2411.11318">pdf</a>, <a href="https://arxiv.org/format/2411.11318">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Syllabus: Portable Curricula for Reinforcement Learning Agents </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Sullivan%2C+R">Ryan Sullivan</a>, <a href="/search/cs?searchtype=author&query=P%C3%A9goud%2C+R">Ryan P茅goud</a>, <a href="/search/cs?searchtype=author&query=Rahmen%2C+A+U">Ameen Ur Rahmen</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xinchen Yang</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+J">Junyun Huang</a>, <a href="/search/cs?searchtype=author&query=Verma%2C+A">Aayush Verma</a>, <a href="/search/cs?searchtype=author&query=Mitra%2C+N">Nistha Mitra</a>, <a href="/search/cs?searchtype=author&query=Dickerson%2C+J+P">John P. Dickerson</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.11318v1-abstract-short" style="display: inline;"> Curriculum learning has been a quiet yet crucial component of many of the high-profile successes of reinforcement learning. Despite this, none of the major reinforcement learning libraries directly support curriculum learning or include curriculum learning implementations. These methods can improve the capabilities and robustness of RL agents, but often require significant, complex changes to agen… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11318v1-abstract-full').style.display = 'inline'; document.getElementById('2411.11318v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.11318v1-abstract-full" style="display: none;"> Curriculum learning has been a quiet yet crucial component of many of the high-profile successes of reinforcement learning. Despite this, none of the major reinforcement learning libraries directly support curriculum learning or include curriculum learning implementations. These methods can improve the capabilities and robustness of RL agents, but often require significant, complex changes to agent training code. We introduce Syllabus, a library for training RL agents with curriculum learning, as a solution to this problem. Syllabus provides a universal API for curriculum learning algorithms, implementations of popular curriculum learning methods, and infrastructure for easily integrating them with distributed training code written in nearly any RL library. Syllabus provides a minimal API for each of the core components of curriculum learning, dramatically simplifying the process of designing new algorithms and applying existing algorithms to new environments. We demonstrate that the same Syllabus code can be used to train agents written in multiple different RL libraries on numerous domains. In doing so, we present the first examples of curriculum learning in NetHack and Neural MMO, two of the premier challenges for single-agent and multi-agent RL respectively, achieving strong results compared to state of the art baselines. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11318v1-abstract-full').style.display = 'none'; document.getElementById('2411.11318v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Preprint</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.11231">arXiv:2411.11231</a> <span> [<a href="https://arxiv.org/pdf/2411.11231">pdf</a>, <a href="https://arxiv.org/format/2411.11231">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> BeautyBank: Encoding Facial Makeup in Latent Space </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lu%2C+Q">Qianwen Lu</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xingchao Yang</a>, <a href="/search/cs?searchtype=author&query=Taketomi%2C+T">Takafumi Taketomi</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.11231v1-abstract-short" style="display: inline;"> The advancement of makeup transfer, editing, and image encoding has demonstrated their effectiveness and superior quality. However, existing makeup works primarily focus on low-dimensional features such as color distributions and patterns, limiting their versatillity across a wide range of makeup applications. Futhermore, existing high-dimensional latent encoding methods mainly target global featu… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11231v1-abstract-full').style.display = 'inline'; document.getElementById('2411.11231v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.11231v1-abstract-full" style="display: none;"> The advancement of makeup transfer, editing, and image encoding has demonstrated their effectiveness and superior quality. However, existing makeup works primarily focus on low-dimensional features such as color distributions and patterns, limiting their versatillity across a wide range of makeup applications. Futhermore, existing high-dimensional latent encoding methods mainly target global features such as structure and style, and are less effective for tasks that require detailed attention to local color and pattern features of makeup. To overcome these limitations, we propose BeautyBank, a novel makeup encoder that disentangles pattern features of bare and makeup faces. Our method encodes makeup features into a high-dimensional space, preserving essential details necessary for makeup reconstruction and broadening the scope of potential makeup research applications. We also propose a Progressive Makeup Tuning (PMT) strategy, specifically designed to enhance the preservation of detailed makeup features while preventing the inclusion of irrelevant attributes. We further explore novel makeup applications, including facial image generation with makeup injection and makeup similarity measure. Extensive empirical experiments validate that our method offers superior task adaptability and holds significant potential for widespread application in various makeup-related fields. Furthermore, to address the lack of large-scale, high-quality paired makeup datasets in the field, we constructed the Bare-Makeup Synthesis Dataset (BMS), comprising 324,000 pairs of 512x512 pixel images of bare and makeup-enhanced faces. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11231v1-abstract-full').style.display = 'none'; document.getElementById('2411.11231v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 17 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.10696">arXiv:2411.10696</a> <span> [<a href="https://arxiv.org/pdf/2411.10696">pdf</a>, <a href="https://arxiv.org/format/2411.10696">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> HELENE: Hessian Layer-wise Clipping and Gradient Annealing for Accelerating Fine-tuning LLM with Zeroth-order Optimization </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhao%2C+H">Huaqin Zhao</a>, <a href="/search/cs?searchtype=author&query=Li%2C+J">Jiaxi Li</a>, <a href="/search/cs?searchtype=author&query=Pan%2C+Y">Yi Pan</a>, <a href="/search/cs?searchtype=author&query=Liang%2C+S">Shizhe Liang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaofeng Yang</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+W">Wei Liu</a>, <a href="/search/cs?searchtype=author&query=Li%2C+X">Xiang Li</a>, <a href="/search/cs?searchtype=author&query=Dou%2C+F">Fei Dou</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+T">Tianming Liu</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+J">Jin Lu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.10696v1-abstract-short" style="display: inline;"> Fine-tuning large language models (LLMs) poses significant memory challenges, as the back-propagation process demands extensive resources, especially with growing model sizes. Recent work, MeZO, addresses this issue using a zeroth-order (ZO) optimization method, which reduces memory consumption by matching the usage to the inference phase. However, MeZO experiences slow convergence due to varying… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.10696v1-abstract-full').style.display = 'inline'; document.getElementById('2411.10696v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.10696v1-abstract-full" style="display: none;"> Fine-tuning large language models (LLMs) poses significant memory challenges, as the back-propagation process demands extensive resources, especially with growing model sizes. Recent work, MeZO, addresses this issue using a zeroth-order (ZO) optimization method, which reduces memory consumption by matching the usage to the inference phase. However, MeZO experiences slow convergence due to varying curvatures across model parameters. To overcome this limitation, we introduce HELENE, a novel scalable and memory-efficient optimizer that integrates annealed A-GNB gradients with a diagonal Hessian estimation and layer-wise clipping, serving as a second-order pre-conditioner. This combination allows for faster and more stable convergence. Our theoretical analysis demonstrates that HELENE improves convergence rates, particularly for models with heterogeneous layer dimensions, by reducing the dependency on the total parameter space dimension. Instead, the method scales with the largest layer dimension, making it highly suitable for modern LLM architectures. Experimental results on RoBERTa-large and OPT-1.3B across multiple tasks show that HELENE achieves up to a 20x speedup compared to MeZO, with average accuracy improvements of 1.5%. Furthermore, HELENE remains compatible with both full parameter tuning and parameter-efficient fine-tuning (PEFT), outperforming several state-of-the-art optimizers. The codes will be released after reviewing. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.10696v1-abstract-full').style.display = 'none'; document.getElementById('2411.10696v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.10507">arXiv:2411.10507</a> <span> [<a href="https://arxiv.org/pdf/2411.10507">pdf</a>, <a href="https://arxiv.org/format/2411.10507">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> RedTest: Towards Measuring Redundancy in Deep Neural Networks Effectively </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lu%2C+Y">Yao Lu</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+P">Peixin Zhang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+J">Jingyi Wang</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+L">Lei Ma</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaoniu Yang</a>, <a href="/search/cs?searchtype=author&query=Xuan%2C+Q">Qi Xuan</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.10507v1-abstract-short" style="display: inline;"> Deep learning has revolutionized computing in many real-world applications, arguably due to its remarkable performance and extreme convenience as an end-to-end solution. However, deep learning models can be costly to train and to use, especially for those large-scale models, making it necessary to optimize the original overly complicated models into smaller ones in scenarios with limited resources… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.10507v1-abstract-full').style.display = 'inline'; document.getElementById('2411.10507v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.10507v1-abstract-full" style="display: none;"> Deep learning has revolutionized computing in many real-world applications, arguably due to its remarkable performance and extreme convenience as an end-to-end solution. However, deep learning models can be costly to train and to use, especially for those large-scale models, making it necessary to optimize the original overly complicated models into smaller ones in scenarios with limited resources such as mobile applications or simply for resource saving. The key question in such model optimization is, how can we effectively identify and measure the redundancy in a deep learning model structure. While several common metrics exist in the popular model optimization techniques to measure the performance of models after optimization, they are not able to quantitatively inform the degree of remaining redundancy. To address the problem, we present a novel testing approach, i.e., RedTest, which proposes a novel testing metric called Model Structural Redundancy Score (MSRS) to quantitatively measure the degree of redundancy in a deep learning model structure. We first show that MSRS is effective in both revealing and assessing the redundancy issues in many state-of-the-art models, which urgently calls for model optimization. Then, we utilize MSRS to assist deep learning model developers in two practical application scenarios: 1) in Neural Architecture Search, we design a novel redundancy-aware algorithm to guide the search for the optimal model structure and demonstrate its effectiveness by comparing it to existing standard NAS practice; 2) in the pruning of large-scale pre-trained models, we prune the redundant layers of pre-trained models with the guidance of layer similarity to derive less redundant ones of much smaller size. Extensive experimental results demonstrate that removing such redundancy has a negligible effect on the model utility. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.10507v1-abstract-full').style.display = 'none'; document.getElementById('2411.10507v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.10332">arXiv:2411.10332</a> <span> [<a href="https://arxiv.org/pdf/2411.10332">pdf</a>, <a href="https://arxiv.org/format/2411.10332">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Number it: Temporal Grounding Videos like Flipping Manga </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wu%2C+Y">Yongliang Wu</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+X">Xinting Hu</a>, <a href="/search/cs?searchtype=author&query=Sun%2C+Y">Yuyang Sun</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+Y">Yizhou Zhou</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+W">Wenbo Zhu</a>, <a href="/search/cs?searchtype=author&query=Rao%2C+F">Fengyun Rao</a>, <a href="/search/cs?searchtype=author&query=Schiele%2C+B">Bernt Schiele</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xu Yang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.10332v1-abstract-short" style="display: inline;"> Video Large Language Models (Vid-LLMs) have made remarkable advancements in comprehending video content for QA dialogue. However, they struggle to extend this visual understanding to tasks requiring precise temporal localization, known as Video Temporal Grounding (VTG). To address this gap, we introduce Number-Prompt (NumPro), a novel method that empowers Vid-LLMs to bridge visual comprehension wi… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.10332v1-abstract-full').style.display = 'inline'; document.getElementById('2411.10332v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.10332v1-abstract-full" style="display: none;"> Video Large Language Models (Vid-LLMs) have made remarkable advancements in comprehending video content for QA dialogue. However, they struggle to extend this visual understanding to tasks requiring precise temporal localization, known as Video Temporal Grounding (VTG). To address this gap, we introduce Number-Prompt (NumPro), a novel method that empowers Vid-LLMs to bridge visual comprehension with temporal grounding by adding unique numerical identifiers to each video frame. Treating a video as a sequence of numbered frame images, NumPro transforms VTG into an intuitive process: flipping through manga panels in sequence. This allows Vid-LLMs to "read" event timelines, accurately linking visual content with corresponding temporal information. Our experiments demonstrate that NumPro significantly boosts VTG performance of top-tier Vid-LLMs without additional computational cost. Furthermore, fine-tuning on a NumPro-enhanced dataset defines a new state-of-the-art for VTG, surpassing previous top-performing methods by up to 6.9\% in mIoU for moment retrieval and 8.5\% in mAP for highlight detection. The code will be available at https://github.com/yongliang-wu/NumPro. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.10332v1-abstract-full').style.display = 'none'; document.getElementById('2411.10332v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">11 pages, 7 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.10279">arXiv:2411.10279</a> <span> [<a href="https://arxiv.org/pdf/2411.10279">pdf</a>, <a href="https://arxiv.org/format/2411.10279">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Cryptography and Security">cs.CR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Lateral Movement Detection via Time-aware Subgraph Classification on Authentication Logs </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhou%2C+J">Jiajun Zhou</a>, <a href="/search/cs?searchtype=author&query=Yao%2C+J">Jiacheng Yao</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+X">Xuanze Chen</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+S">Shanqing Yu</a>, <a href="/search/cs?searchtype=author&query=Xuan%2C+Q">Qi Xuan</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaoniu Yang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.10279v1-abstract-short" style="display: inline;"> Lateral movement is a crucial component of advanced persistent threat (APT) attacks in networks. Attackers exploit security vulnerabilities in internal networks or IoT devices, expanding their control after initial infiltration to steal sensitive data or carry out other malicious activities, posing a serious threat to system security. Existing research suggests that attackers generally employ seem… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.10279v1-abstract-full').style.display = 'inline'; document.getElementById('2411.10279v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.10279v1-abstract-full" style="display: none;"> Lateral movement is a crucial component of advanced persistent threat (APT) attacks in networks. Attackers exploit security vulnerabilities in internal networks or IoT devices, expanding their control after initial infiltration to steal sensitive data or carry out other malicious activities, posing a serious threat to system security. Existing research suggests that attackers generally employ seemingly unrelated operations to mask their malicious intentions, thereby evading existing lateral movement detection methods and hiding their intrusion traces. In this regard, we analyze host authentication log data from a graph perspective and propose a multi-scale lateral movement detection framework called LMDetect. The main workflow of this framework proceeds as follows: 1) Construct a heterogeneous multigraph from host authentication log data to strengthen the correlations among internal system entities; 2) Design a time-aware subgraph generator to extract subgraphs centered on authentication events from the heterogeneous authentication multigraph; 3) Design a multi-scale attention encoder that leverages both local and global attention to capture hidden anomalous behavior patterns in the authentication subgraphs, thereby achieving lateral movement detection. Extensive experiments on two real-world authentication log datasets demonstrate the effectiveness and superiority of our framework in detecting lateral movement behaviors. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.10279v1-abstract-full').style.display = 'none'; document.getElementById('2411.10279v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.09858">arXiv:2411.09858</a> <span> [<a href="https://arxiv.org/pdf/2411.09858">pdf</a>, <a href="https://arxiv.org/format/2411.09858">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Masked Image Contrastive Learning for Efficient Visual Conceptual Pre-training </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaoyu Yang</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+L">Lijian Xu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.09858v1-abstract-short" style="display: inline;"> This paper proposes a scalable and straightforward pre-training paradigm for efficient visual conceptual representation called masked image contrastive learning (MiCL). Our MiCL approach is simple: we randomly mask patches to generate different views within an image and contrast them among a mini-batch of images. The core idea behind MiCL consists of two designs. First, masked tokens have the pote… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.09858v1-abstract-full').style.display = 'inline'; document.getElementById('2411.09858v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.09858v1-abstract-full" style="display: none;"> This paper proposes a scalable and straightforward pre-training paradigm for efficient visual conceptual representation called masked image contrastive learning (MiCL). Our MiCL approach is simple: we randomly mask patches to generate different views within an image and contrast them among a mini-batch of images. The core idea behind MiCL consists of two designs. First, masked tokens have the potential to significantly diminish the conceptual redundancy inherent in images, and create distinct views with substantial fine-grained differences on the semantic concept level instead of the instance level. Second, contrastive learning is adept at extracting high-level semantic conceptual features during the pre-training, circumventing the high-frequency interference and additional costs associated with image reconstruction. Importantly, MiCL learns highly semantic conceptual representations efficiently without relying on hand-crafted data augmentations or additional auxiliary modules. Empirically, MiCL demonstrates high scalability with Vision Transformers, as the ViT-L/16 can complete pre-training in 133 hours using only 4 A100 GPUs, achieving 85.8% accuracy in downstream fine-tuning tasks. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.09858v1-abstract-full').style.display = 'none'; document.getElementById('2411.09858v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 14 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">10 pages</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> I.2.10 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.09644">arXiv:2411.09644</a> <span> [<a href="https://arxiv.org/pdf/2411.09644">pdf</a>, <a href="https://arxiv.org/format/2411.09644">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Optimization and Control">math.OC</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Numerical Analysis">math.NA</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Probability">math.PR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computational Finance">q-fin.CP</span> </div> </div> <p class="title is-5 mathjax"> Neural Operators Can Play Dynamic Stackelberg Games </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Alvarez%2C+G">Guillermo Alvarez</a>, <a href="/search/cs?searchtype=author&query=Ekren%2C+I">Ibrahim Ekren</a>, <a href="/search/cs?searchtype=author&query=Kratsios%2C+A">Anastasis Kratsios</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xuwei Yang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.09644v1-abstract-short" style="display: inline;"> Dynamic Stackelberg games are a broad class of two-player games in which the leader acts first, and the follower chooses a response strategy to the leader's strategy. Unfortunately, only stylized Stackelberg games are explicitly solvable since the follower's best-response operator (as a function of the control of the leader) is typically analytically intractable. This paper addresses this issue by… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.09644v1-abstract-full').style.display = 'inline'; document.getElementById('2411.09644v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.09644v1-abstract-full" style="display: none;"> Dynamic Stackelberg games are a broad class of two-player games in which the leader acts first, and the follower chooses a response strategy to the leader's strategy. Unfortunately, only stylized Stackelberg games are explicitly solvable since the follower's best-response operator (as a function of the control of the leader) is typically analytically intractable. This paper addresses this issue by showing that the \textit{follower's best-response operator} can be approximately implemented by an \textit{attention-based neural operator}, uniformly on compact subsets of adapted open-loop controls for the leader. We further show that the value of the Stackelberg game where the follower uses the approximate best-response operator approximates the value of the original Stackelberg game. Our main result is obtained using our universal approximation theorem for attention-based neural operators between spaces of square-integrable adapted stochastic processes, as well as stability results for a general class of Stackelberg games. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.09644v1-abstract-full').style.display = 'none'; document.getElementById('2411.09644v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 14 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.09422">arXiv:2411.09422</a> <span> [<a href="https://arxiv.org/pdf/2411.09422">pdf</a>, <a href="https://arxiv.org/format/2411.09422">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> OpenLS-DGF: An Adaptive Open-Source Dataset Generation Framework for Machine Learning Tasks in Logic Synthesis </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ni%2C+L">Liwei Ni</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+R">Rui Wang</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+M">Miao Liu</a>, <a href="/search/cs?searchtype=author&query=Meng%2C+X">Xingyu Meng</a>, <a href="/search/cs?searchtype=author&query=Lin%2C+X">Xiaoze Lin</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+J">Junfeng Liu</a>, <a href="/search/cs?searchtype=author&query=Luo%2C+G">Guojie Luo</a>, <a href="/search/cs?searchtype=author&query=Chu%2C+Z">Zhufei Chu</a>, <a href="/search/cs?searchtype=author&query=Qian%2C+W">Weikang Qian</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaoyan Yang</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+B">Biwei Xie</a>, <a href="/search/cs?searchtype=author&query=Li%2C+X">Xingquan Li</a>, <a href="/search/cs?searchtype=author&query=Li%2C+H">Huawei Li</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.09422v2-abstract-short" style="display: inline;"> This paper introduces OpenLS-DGF, an adaptive logic synthesis dataset generation framework, to enhance machine learning~(ML) applications within the logic synthesis process. Previous dataset generation flows were tailored for specific tasks or lacked integrated machine learning capabilities. While OpenLS-DGF supports various machine learning tasks by encapsulating the three fundamental steps of lo… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.09422v2-abstract-full').style.display = 'inline'; document.getElementById('2411.09422v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.09422v2-abstract-full" style="display: none;"> This paper introduces OpenLS-DGF, an adaptive logic synthesis dataset generation framework, to enhance machine learning~(ML) applications within the logic synthesis process. Previous dataset generation flows were tailored for specific tasks or lacked integrated machine learning capabilities. While OpenLS-DGF supports various machine learning tasks by encapsulating the three fundamental steps of logic synthesis: Boolean representation, logic optimization, and technology mapping. It preserves the original information in both Verilog and machine-learning-friendly GraphML formats. The verilog files offer semi-customizable capabilities, enabling researchers to insert additional steps and incrementally refine the generated dataset. Furthermore, OpenLS-DGF includes an adaptive circuit engine that facilitates the final dataset management and downstream tasks. The generated OpenLS-D-v1 dataset comprises 46 combinational designs from established benchmarks, totaling over 966,000 Boolean circuits. OpenLS-D-v1 supports integrating new data features, making it more versatile for new challenges. This paper demonstrates the versatility of OpenLS-D-v1 through four distinct downstream tasks: circuit classification, circuit ranking, quality of results (QoR) prediction, and probability prediction. Each task is chosen to represent essential steps of logic synthesis, and the experimental results show the generated dataset from OpenLS-DGF achieves prominent diversity and applicability. The source code and datasets are available at https://github.com/Logic-Factory/ACE/blob/master/OpenLS-DGF/readme.md. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.09422v2-abstract-full').style.display = 'none'; document.getElementById('2411.09422v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 16 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">14 pages</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.09189">arXiv:2411.09189</a> <span> [<a href="https://arxiv.org/pdf/2411.09189">pdf</a>, <a href="https://arxiv.org/format/2411.09189">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Sound">cs.SD</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Audio and Speech Processing">eess.AS</span> </div> </div> <p class="title is-5 mathjax"> Improvement and Implementation of a Speech Emotion Recognition Model Based on Dual-Layer LSTM </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaoran Yang</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+S">Shuhan Yu</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+W">Wenxi Xu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.09189v1-abstract-short" style="display: inline;"> This paper builds upon an existing speech emotion recognition model by adding an additional LSTM layer to improve the accuracy and processing efficiency of emotion recognition from audio data. By capturing the long-term dependencies within audio sequences through a dual-layer LSTM network, the model can recognize and classify complex emotional patterns more accurately. Experiments conducted on the… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.09189v1-abstract-full').style.display = 'inline'; document.getElementById('2411.09189v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.09189v1-abstract-full" style="display: none;"> This paper builds upon an existing speech emotion recognition model by adding an additional LSTM layer to improve the accuracy and processing efficiency of emotion recognition from audio data. By capturing the long-term dependencies within audio sequences through a dual-layer LSTM network, the model can recognize and classify complex emotional patterns more accurately. Experiments conducted on the RAVDESS dataset validated this approach, showing that the modified dual layer LSTM model improves accuracy by 2% compared to the single-layer LSTM while significantly reducing recognition latency, thereby enhancing real-time performance. These results indicate that the dual-layer LSTM architecture is highly suitable for handling emotional features with long-term dependencies, providing a viable optimization for speech emotion recognition systems. This research provides a reference for practical applications in fields like intelligent customer service, sentiment analysis and human-computer interaction. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.09189v1-abstract-full').style.display = 'none'; document.getElementById('2411.09189v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 14 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.08534">arXiv:2411.08534</a> <span> [<a href="https://arxiv.org/pdf/2411.08534">pdf</a>, <a href="https://arxiv.org/format/2411.08534">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> </div> </div> <p class="title is-5 mathjax"> Neural Topic Modeling with Large Language Models in the Loop </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaohao Yang</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+H">He Zhao</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+W">Weijie Xu</a>, <a href="/search/cs?searchtype=author&query=Qi%2C+Y">Yuanyuan Qi</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+J">Jueqing Lu</a>, <a href="/search/cs?searchtype=author&query=Phung%2C+D">Dinh Phung</a>, <a href="/search/cs?searchtype=author&query=Du%2C+L">Lan Du</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.08534v1-abstract-short" style="display: inline;"> Topic modeling is a fundamental task in natural language processing, allowing the discovery of latent thematic structures in text corpora. While Large Language Models (LLMs) have demonstrated promising capabilities in topic discovery, their direct application to topic modeling suffers from issues such as incomplete topic coverage, misalignment of topics, and inefficiency. To address these limitati… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.08534v1-abstract-full').style.display = 'inline'; document.getElementById('2411.08534v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.08534v1-abstract-full" style="display: none;"> Topic modeling is a fundamental task in natural language processing, allowing the discovery of latent thematic structures in text corpora. While Large Language Models (LLMs) have demonstrated promising capabilities in topic discovery, their direct application to topic modeling suffers from issues such as incomplete topic coverage, misalignment of topics, and inefficiency. To address these limitations, we propose LLM-ITL, a novel LLM-in-the-loop framework that integrates LLMs with many existing Neural Topic Models (NTMs). In LLM-ITL, global topics and document representations are learned through the NTM, while an LLM refines the topics via a confidence-weighted Optimal Transport (OT)-based alignment objective. This process enhances the interpretability and coherence of the learned topics, while maintaining the efficiency of NTMs. Extensive experiments demonstrate that LLM-ITL can help NTMs significantly improve their topic interpretability while maintaining the quality of document representation. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.08534v1-abstract-full').style.display = 'none'; document.getElementById('2411.08534v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 13 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.08453">arXiv:2411.08453</a> <span> [<a href="https://arxiv.org/pdf/2411.08453">pdf</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Biomass phenotyping of oilseed rape through UAV multi-view oblique imaging with 3DGS and SAM model </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shen%2C+Y">Yutao Shen</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+H">Hongyu Zhou</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xin Yang</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+X">Xuqi Lu</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+Z">Ziyue Guo</a>, <a href="/search/cs?searchtype=author&query=Jiang%2C+L">Lixi Jiang</a>, <a href="/search/cs?searchtype=author&query=He%2C+Y">Yong He</a>, <a href="/search/cs?searchtype=author&query=Cen%2C+H">Haiyan Cen</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.08453v1-abstract-short" style="display: inline;"> Biomass estimation of oilseed rape is crucial for optimizing crop productivity and breeding strategies. While UAV-based imaging has advanced high-throughput phenotyping, current methods often rely on orthophoto images, which struggle with overlapping leaves and incomplete structural information in complex field environments. This study integrates 3D Gaussian Splatting (3DGS) with the Segment Anyth… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.08453v1-abstract-full').style.display = 'inline'; document.getElementById('2411.08453v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.08453v1-abstract-full" style="display: none;"> Biomass estimation of oilseed rape is crucial for optimizing crop productivity and breeding strategies. While UAV-based imaging has advanced high-throughput phenotyping, current methods often rely on orthophoto images, which struggle with overlapping leaves and incomplete structural information in complex field environments. This study integrates 3D Gaussian Splatting (3DGS) with the Segment Anything Model (SAM) for precise 3D reconstruction and biomass estimation of oilseed rape. UAV multi-view oblique images from 36 angles were used to perform 3D reconstruction, with the SAM module enhancing point cloud segmentation. The segmented point clouds were then converted into point cloud volumes, which were fitted to ground-measured biomass using linear regression. The results showed that 3DGS (7k and 30k iterations) provided high accuracy, with peak signal-to-noise ratios (PSNR) of 27.43 and 29.53 and training times of 7 and 49 minutes, respectively. This performance exceeded that of structure from motion (SfM) and mipmap Neural Radiance Fields (Mip-NeRF), demonstrating superior efficiency. The SAM module achieved high segmentation accuracy, with a mean intersection over union (mIoU) of 0.961 and an F1-score of 0.980. Additionally, a comparison of biomass extraction models found the point cloud volume model to be the most accurate, with an determination coefficient (R2) of 0.976, root mean square error (RMSE) of 2.92 g/plant, and mean absolute percentage error (MAPE) of 6.81%, outperforming both the plot crop volume and individual crop volume models. This study highlights the potential of combining 3DGS with multi-view UAV imaging for improved biomass phenotyping. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.08453v1-abstract-full').style.display = 'none'; document.getElementById('2411.08453v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 13 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.08433">arXiv:2411.08433</a> <span> [<a href="https://arxiv.org/pdf/2411.08433">pdf</a>, <a href="https://arxiv.org/format/2411.08433">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Robotics">cs.RO</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> 3D Multi-Object Tracking with Semi-Supervised GRU-Kalman Filter </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wang%2C+X">Xiaoxiang Wang</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+J">Jiaxin Liu</a>, <a href="/search/cs?searchtype=author&query=Feng%2C+M">Miaojie Feng</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Z">Zhaoxing Zhang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xin Yang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.08433v1-abstract-short" style="display: inline;"> 3D Multi-Object Tracking (MOT), a fundamental component of environmental perception, is essential for intelligent systems like autonomous driving and robotic sensing. Although Tracking-by-Detection frameworks have demonstrated excellent performance in recent years, their application in real-world scenarios faces significant challenges. Object movement in complex environments is often highly nonlin… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.08433v1-abstract-full').style.display = 'inline'; document.getElementById('2411.08433v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.08433v1-abstract-full" style="display: none;"> 3D Multi-Object Tracking (MOT), a fundamental component of environmental perception, is essential for intelligent systems like autonomous driving and robotic sensing. Although Tracking-by-Detection frameworks have demonstrated excellent performance in recent years, their application in real-world scenarios faces significant challenges. Object movement in complex environments is often highly nonlinear, while existing methods typically rely on linear approximations of motion. Furthermore, system noise is frequently modeled as a Gaussian distribution, which fails to capture the true complexity of the noise dynamics. These oversimplified modeling assumptions can lead to significant reductions in tracking precision. To address this, we propose a GRU-based MOT method, which introduces a learnable Kalman filter into the motion module. This approach is able to learn object motion characteristics through data-driven learning, thereby avoiding the need for manual model design and model error. At the same time, to avoid abnormal supervision caused by the wrong association between annotations and trajectories, we design a semi-supervised learning strategy to accelerate the convergence speed and improve the robustness of the model. Evaluation experiment on the nuScenes and Argoverse2 datasets demonstrates that our system exhibits superior performance and significant potential compared to traditional TBD methods. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.08433v1-abstract-full').style.display = 'none'; document.getElementById('2411.08433v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 13 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.07611">arXiv:2411.07611</a> <span> [<a href="https://arxiv.org/pdf/2411.07611">pdf</a>, <a href="https://arxiv.org/format/2411.07611">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Multimodal Clinical Reasoning through Knowledge-augmented Rationale Generation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Niu%2C+S">Shuai Niu</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+J">Jing Ma</a>, <a href="/search/cs?searchtype=author&query=Bai%2C+L">Liang Bai</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Z">Zhihua Wang</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+Y">Yida Xu</a>, <a href="/search/cs?searchtype=author&query=Song%2C+Y">Yunya Song</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xian Yang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.07611v1-abstract-short" style="display: inline;"> Clinical rationales play a pivotal role in accurate disease diagnosis; however, many models predominantly use discriminative methods and overlook the importance of generating supportive rationales. Rationale distillation is a process that transfers knowledge from large language models (LLMs) to smaller language models (SLMs), thereby enhancing the latter's ability to break down complex tasks. Desp… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.07611v1-abstract-full').style.display = 'inline'; document.getElementById('2411.07611v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.07611v1-abstract-full" style="display: none;"> Clinical rationales play a pivotal role in accurate disease diagnosis; however, many models predominantly use discriminative methods and overlook the importance of generating supportive rationales. Rationale distillation is a process that transfers knowledge from large language models (LLMs) to smaller language models (SLMs), thereby enhancing the latter's ability to break down complex tasks. Despite its benefits, rationale distillation alone is inadequate for addressing domain knowledge limitations in tasks requiring specialized expertise, such as disease diagnosis. Effectively embedding domain knowledge in SLMs poses a significant challenge. While current LLMs are primarily geared toward processing textual data, multimodal LLMs that incorporate time series data, especially electronic health records (EHRs), are still evolving. To tackle these limitations, we introduce ClinRaGen, an SLM optimized for multimodal rationale generation in disease diagnosis. ClinRaGen incorporates a unique knowledge-augmented attention mechanism to merge domain knowledge with time series EHR data, utilizing a stepwise rationale distillation strategy to produce both textual and time series-based clinical rationales. Our evaluations show that ClinRaGen markedly improves the SLM's capability to interpret multimodal EHR data and generate accurate clinical rationales, supporting more reliable disease diagnosis, advancing LLM applications in healthcare, and narrowing the performance divide between LLMs and SLMs. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.07611v1-abstract-full').style.display = 'none'; document.getElementById('2411.07611v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 12 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">11 pages. 4 figures</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> I.2.7 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.07574">arXiv:2411.07574</a> <span> [<a href="https://arxiv.org/pdf/2411.07574">pdf</a>, <a href="https://arxiv.org/format/2411.07574">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Disentangling Tabular Data towards Better One-Class Anomaly Detection </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ye%2C+J">Jianan Ye</a>, <a href="/search/cs?searchtype=author&query=Tan%2C+Z">Zhaorui Tan</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yijie Hu</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xi Yang</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+G">Guangliang Cheng</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+K">Kaizhu Huang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.07574v1-abstract-short" style="display: inline;"> Tabular anomaly detection under the one-class classification setting poses a significant challenge, as it involves accurately conceptualizing "normal" derived exclusively from a single category to discern anomalies from normal data variations. Capturing the intrinsic correlation among attributes within normal samples presents one promising method for learning the concept. To do so, the most recent… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.07574v1-abstract-full').style.display = 'inline'; document.getElementById('2411.07574v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.07574v1-abstract-full" style="display: none;"> Tabular anomaly detection under the one-class classification setting poses a significant challenge, as it involves accurately conceptualizing "normal" derived exclusively from a single category to discern anomalies from normal data variations. Capturing the intrinsic correlation among attributes within normal samples presents one promising method for learning the concept. To do so, the most recent effort relies on a learnable mask strategy with a reconstruction task. However, this wisdom may suffer from the risk of producing uniform masks, i.e., essentially nothing is masked, leading to less effective correlation learning. To address this issue, we presume that attributes related to others in normal samples can be divided into two non-overlapping and correlated subsets, defined as CorrSets, to capture the intrinsic correlation effectively. Accordingly, we introduce an innovative method that disentangles CorrSets from normal tabular data. To our knowledge, this is a pioneering effort to apply the concept of disentanglement for one-class anomaly detection on tabular data. Extensive experiments on 20 tabular datasets show that our method substantially outperforms the state-of-the-art methods and leads to an average performance improvement of 6.1% on AUC-PR and 2.1% on AUC-ROC. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.07574v1-abstract-full').style.display = 'none'; document.getElementById('2411.07574v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 12 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.07025">arXiv:2411.07025</a> <span> [<a href="https://arxiv.org/pdf/2411.07025">pdf</a>, <a href="https://arxiv.org/format/2411.07025">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Graphics">cs.GR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Scaling Mesh Generation via Compressive Tokenization </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Weng%2C+H">Haohan Weng</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+Z">Zibo Zhao</a>, <a href="/search/cs?searchtype=author&query=Lei%2C+B">Biwen Lei</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xianghui Yang</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+J">Jian Liu</a>, <a href="/search/cs?searchtype=author&query=Lai%2C+Z">Zeqiang Lai</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Z">Zhuo Chen</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Y">Yuhong Liu</a>, <a href="/search/cs?searchtype=author&query=Jiang%2C+J">Jie Jiang</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+C">Chunchao Guo</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+T">Tong Zhang</a>, <a href="/search/cs?searchtype=author&query=Gao%2C+S">Shenghua Gao</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+C+L+P">C. L. Philip Chen</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.07025v1-abstract-short" style="display: inline;"> We propose a compressive yet effective mesh representation, Blocked and Patchified Tokenization (BPT), facilitating the generation of meshes exceeding 8k faces. BPT compresses mesh sequences by employing block-wise indexing and patch aggregation, reducing their length by approximately 75\% compared to the original sequences. This compression milestone unlocks the potential to utilize mesh data wit… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.07025v1-abstract-full').style.display = 'inline'; document.getElementById('2411.07025v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.07025v1-abstract-full" style="display: none;"> We propose a compressive yet effective mesh representation, Blocked and Patchified Tokenization (BPT), facilitating the generation of meshes exceeding 8k faces. BPT compresses mesh sequences by employing block-wise indexing and patch aggregation, reducing their length by approximately 75\% compared to the original sequences. This compression milestone unlocks the potential to utilize mesh data with significantly more faces, thereby enhancing detail richness and improving generation robustness. Empowered with the BPT, we have built a foundation mesh generative model training on scaled mesh data to support flexible control for point clouds and images. Our model demonstrates the capability to generate meshes with intricate details and accurate topology, achieving SoTA performance on mesh generation and reaching the level for direct product usage. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.07025v1-abstract-full').style.display = 'none'; document.getElementById('2411.07025v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 11 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Homepage: https://whaohan.github.io/bpt , Code: https://github.com/whaohan/bpt</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.06920">arXiv:2411.06920</a> <span> [<a href="https://arxiv.org/pdf/2411.06920">pdf</a>, <a href="https://arxiv.org/format/2411.06920">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Robotics">cs.RO</span> </div> </div> <p class="title is-5 mathjax"> Safe Planner: Empowering Safety Awareness in Large Pre-Trained Models for Robot Task Planning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Li%2C+S">Siyuan Li</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+Z">Zhe Ma</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+F">Feifan Liu</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+J">Jiani Lu</a>, <a href="/search/cs?searchtype=author&query=Xiao%2C+Q">Qinqin Xiao</a>, <a href="/search/cs?searchtype=author&query=Sun%2C+K">Kewu Sun</a>, <a href="/search/cs?searchtype=author&query=Cui%2C+L">Lingfei Cui</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xirui Yang</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+P">Peng Liu</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+X">Xun Wang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.06920v1-abstract-short" style="display: inline;"> Robot task planning is an important problem for autonomous robots in long-horizon challenging tasks. As large pre-trained models have demonstrated superior planning ability, recent research investigates utilizing large models to achieve autonomous planning for robots in diverse tasks. However, since the large models are pre-trained with Internet data and lack the knowledge of real task scenes, lar… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.06920v1-abstract-full').style.display = 'inline'; document.getElementById('2411.06920v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.06920v1-abstract-full" style="display: none;"> Robot task planning is an important problem for autonomous robots in long-horizon challenging tasks. As large pre-trained models have demonstrated superior planning ability, recent research investigates utilizing large models to achieve autonomous planning for robots in diverse tasks. However, since the large models are pre-trained with Internet data and lack the knowledge of real task scenes, large models as planners may make unsafe decisions that hurt the robots and the surrounding environments. To solve this challenge, we propose a novel Safe Planner framework, which empowers safety awareness in large pre-trained models to accomplish safe and executable planning. In this framework, we develop a safety prediction module to guide the high-level large model planner, and this safety module trained in a simulator can be effectively transferred to real-world tasks. The proposed Safe Planner framework is evaluated on both simulated environments and real robots. The experiment results demonstrate that Safe Planner not only achieves state-of-the-art task success rates, but also substantially improves safety during task execution. The experiment videos are shown in https://sites.google.com/view/safeplanner . <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.06920v1-abstract-full').style.display = 'none'; document.getElementById('2411.06920v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 11 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">9 pages, 6 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.06852">arXiv:2411.06852</a> <span> [<a href="https://arxiv.org/pdf/2411.06852">pdf</a>, <a href="https://arxiv.org/format/2411.06852">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Evaluating Large Language Models on Financial Report Summarization: An Empirical Study </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xinqi Yang</a>, <a href="/search/cs?searchtype=author&query=Zang%2C+S">Scott Zang</a>, <a href="/search/cs?searchtype=author&query=Ren%2C+Y">Yong Ren</a>, <a href="/search/cs?searchtype=author&query=Peng%2C+D">Dingjie Peng</a>, <a href="/search/cs?searchtype=author&query=Wen%2C+Z">Zheng Wen</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.06852v1-abstract-short" style="display: inline;"> In recent years, Large Language Models (LLMs) have demonstrated remarkable versatility across various applications, including natural language understanding, domain-specific knowledge tasks, etc. However, applying LLMs to complex, high-stakes domains like finance requires rigorous evaluation to ensure reliability, accuracy, and compliance with industry standards. To address this need, we conduct a… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.06852v1-abstract-full').style.display = 'inline'; document.getElementById('2411.06852v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.06852v1-abstract-full" style="display: none;"> In recent years, Large Language Models (LLMs) have demonstrated remarkable versatility across various applications, including natural language understanding, domain-specific knowledge tasks, etc. However, applying LLMs to complex, high-stakes domains like finance requires rigorous evaluation to ensure reliability, accuracy, and compliance with industry standards. To address this need, we conduct a comprehensive and comparative study on three state-of-the-art LLMs, GLM-4, Mistral-NeMo, and LLaMA3.1, focusing on their effectiveness in generating automated financial reports. Our primary motivation is to explore how these models can be harnessed within finance, a field demanding precision, contextual relevance, and robustness against erroneous or misleading information. By examining each model's capabilities, we aim to provide an insightful assessment of their strengths and limitations. Our paper offers benchmarks for financial report analysis, encompassing proposed metrics such as ROUGE-1, BERT Score, and LLM Score. We introduce an innovative evaluation framework that integrates both quantitative metrics (e.g., precision, recall) and qualitative analyses (e.g., contextual fit, consistency) to provide a holistic view of each model's output quality. Additionally, we make our financial dataset publicly available, inviting researchers and practitioners to leverage, scrutinize, and enhance our findings through broader community engagement and collaborative improvement. Our dataset is available on huggingface. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.06852v1-abstract-full').style.display = 'none'; document.getElementById('2411.06852v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 11 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.06666">arXiv:2411.06666</a> <span> [<a href="https://arxiv.org/pdf/2411.06666">pdf</a>, <a href="https://arxiv.org/format/2411.06666">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Adversarial Detection with a Dynamically Stable System </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Long%2C+X">Xiaowei Long</a>, <a href="/search/cs?searchtype=author&query=Lin%2C+J">Jie Lin</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiangyuan Yang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.06666v1-abstract-short" style="display: inline;"> Adversarial detection is designed to identify and reject maliciously crafted adversarial examples(AEs) which are generated to disrupt the classification of target models. Presently, various input transformation-based methods have been developed on adversarial example detection, which typically rely on empirical experience and lead to unreliability against new attacks. To address this issue, we… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.06666v1-abstract-full').style.display = 'inline'; document.getElementById('2411.06666v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.06666v1-abstract-full" style="display: none;"> Adversarial detection is designed to identify and reject maliciously crafted adversarial examples(AEs) which are generated to disrupt the classification of target models. Presently, various input transformation-based methods have been developed on adversarial example detection, which typically rely on empirical experience and lead to unreliability against new attacks. To address this issue, we propose and conduct a Dynamically Stable System (DSS), which can effectively detect the adversarial examples from normal examples according to the stability of input examples. Particularly, in our paper, the generation of adversarial examples is considered as the perturbation process of a Lyapunov dynamic system, and we propose an example stability mechanism, in which a novel control term is added in adversarial example generation to ensure that the normal examples can achieve dynamic stability while the adversarial examples cannot achieve the stability. Then, based on the proposed example stability mechanism, a Dynamically Stable System (DSS) is proposed, which can utilize the disruption and restoration actions to determine the stability of input examples and detect the adversarial examples through changes in the stability of the input examples. In comparison with existing methods in three benchmark datasets(MNIST, CIFAR10, and CIFAR100), our evaluation results show that our proposed DSS can achieve ROC-AUC values of 99.83%, 97.81% and 94.47%, surpassing the state-of-the-art(SOTA) values of 97.35%, 91.10% and 93.49% in the other 7 methods. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.06666v1-abstract-full').style.display = 'none'; document.getElementById('2411.06666v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 10 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.06106">arXiv:2411.06106</a> <span> [<a href="https://arxiv.org/pdf/2411.06106">pdf</a>, <a href="https://arxiv.org/format/2411.06106">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Personalize to generalize: Towards a universal medical multi-modality generalization through personalization </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Tan%2C+Z">Zhaorui Tan</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xi Yang</a>, <a href="/search/cs?searchtype=author&query=Pan%2C+T">Tan Pan</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+T">Tianyi Liu</a>, <a href="/search/cs?searchtype=author&query=Jiang%2C+C">Chen Jiang</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+X">Xin Guo</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Q">Qiufeng Wang</a>, <a href="/search/cs?searchtype=author&query=Nguyen%2C+A">Anh Nguyen</a>, <a href="/search/cs?searchtype=author&query=Qi%2C+Y">Yuan Qi</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+K">Kaizhu Huang</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+Y">Yuan Cheng</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.06106v2-abstract-short" style="display: inline;"> The differences among medical imaging modalities, driven by distinct underlying principles, pose significant challenges for generalization in multi-modal medical tasks. Beyond modality gaps, individual variations, such as differences in organ size and metabolic rate, further impede a model's ability to generalize effectively across both modalities and diverse populations. Despite the importance of… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.06106v2-abstract-full').style.display = 'inline'; document.getElementById('2411.06106v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.06106v2-abstract-full" style="display: none;"> The differences among medical imaging modalities, driven by distinct underlying principles, pose significant challenges for generalization in multi-modal medical tasks. Beyond modality gaps, individual variations, such as differences in organ size and metabolic rate, further impede a model's ability to generalize effectively across both modalities and diverse populations. Despite the importance of personalization, existing approaches to multi-modal generalization often neglect individual differences, focusing solely on common anatomical features. This limitation may result in weakened generalization in various medical tasks. In this paper, we unveil that personalization is critical for multi-modal generalization. Specifically, we propose an approach to achieve personalized generalization through approximating the underlying personalized invariant representation ${X}_h$ across various modalities by leveraging individual-level constraints and a learnable biological prior. We validate the feasibility and benefits of learning a personalized ${X}_h$, showing that this representation is highly generalizable and transferable across various multi-modal medical tasks. Extensive experimental results consistently show that the additionally incorporated personalization significantly improves performance and generalization across diverse scenarios, confirming its effectiveness. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.06106v2-abstract-full').style.display = 'none'; document.getElementById('2411.06106v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 12 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 9 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.06102">arXiv:2411.06102</a> <span> [<a href="https://arxiv.org/pdf/2411.06102">pdf</a>, <a href="https://arxiv.org/format/2411.06102">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Databases">cs.DB</span> </div> </div> <p class="title is-5 mathjax"> SiriusBI: Building End-to-End Business Intelligence Enhanced by Large Language Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jiang%2C+J">Jie Jiang</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+H">Haining Xie</a>, <a href="/search/cs?searchtype=author&query=Shen%2C+Y">Yu Shen</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Z">Zihan Zhang</a>, <a href="/search/cs?searchtype=author&query=Lei%2C+M">Meng Lei</a>, <a href="/search/cs?searchtype=author&query=Zheng%2C+Y">Yifeng Zheng</a>, <a href="/search/cs?searchtype=author&query=Fang%2C+Y">Yide Fang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+C">Chunyou Li</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+D">Danqing Huang</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+W">Wentao Zhang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+Y">Yang Li</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaofeng Yang</a>, <a href="/search/cs?searchtype=author&query=Cui%2C+B">Bin Cui</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+P">Peng Chen</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.06102v1-abstract-short" style="display: inline;"> The rapid advancement of AI technologies, particularly Large Language Models (LLMs), is establishing a new paradigm for Business Intelligence (BI). Despite the emergence of pioneering work in enhancing BI systems with LLMs, we have identified the following three issues when deployed in real industrial scenarios: interaction limitations, performance bottlenecks, and functionality deficiencies. In… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.06102v1-abstract-full').style.display = 'inline'; document.getElementById('2411.06102v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.06102v1-abstract-full" style="display: none;"> The rapid advancement of AI technologies, particularly Large Language Models (LLMs), is establishing a new paradigm for Business Intelligence (BI). Despite the emergence of pioneering work in enhancing BI systems with LLMs, we have identified the following three issues when deployed in real industrial scenarios: interaction limitations, performance bottlenecks, and functionality deficiencies. In this paper, we present SiriusBI, an end-to-end business intelligence system that is designed to address the three issues simultaneously. First, we propose an intelligent and application-oriented module called multi-round dialogue with querying, which aims to overcome the prevalent interaction limitations in current BI solutions. Next, to mitigate the performance bottlenecks caused by scenario migration, we introduce two SQL generation methods that strike a balance between accuracy and deployment costs. Finally, to tackle the practical challenges posed by functionality deficiencies, we develop an end-to-end workflow that covers the entire BI process, ensuring that SiriusBI delivers a robust and complete set of functionalities. As an independent cloud service in Tencent's data platform, SiriusBI has been applied across Tencent's finance, advertising, and cloud sectors, providing services to dozens of enterprise clients. Experiments on real-world datasets and practical applications in industrial BI scenarios demonstrate the practicality and effectiveness of SiriusBI. Remarkably, SiriusBI achieves remarkable accuracy rates of 97% in SQL generation for Tencent Finance, 89% for Tencent Advertisement, and 91% for Tencent Cloud. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.06102v1-abstract-full').style.display = 'none'; document.getElementById('2411.06102v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 9 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">14 pages, 5figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.05945">arXiv:2411.05945</a> <span> [<a href="https://arxiv.org/pdf/2411.05945">pdf</a>, <a href="https://arxiv.org/format/2411.05945">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Multiagent Systems">cs.MA</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Audio and Speech Processing">eess.AS</span> </div> </div> <p class="title is-5 mathjax"> NeKo: Toward Post Recognition Generative Correction Large Language Models with Task-Oriented Experts </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lin%2C+Y">Yen-Ting Lin</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+C+H">Chao-Han Huck Yang</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Z">Zhehuai Chen</a>, <a href="/search/cs?searchtype=author&query=Zelasko%2C+P">Piotr Zelasko</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xuesong Yang</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Z">Zih-Ching Chen</a>, <a href="/search/cs?searchtype=author&query=Puvvada%2C+K+C">Krishna C Puvvada</a>, <a href="/search/cs?searchtype=author&query=Fu%2C+S">Szu-Wei Fu</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+K">Ke Hu</a>, <a href="/search/cs?searchtype=author&query=Chiu%2C+J+W">Jun Wei Chiu</a>, <a href="/search/cs?searchtype=author&query=Balam%2C+J">Jagadeesh Balam</a>, <a href="/search/cs?searchtype=author&query=Ginsburg%2C+B">Boris Ginsburg</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y+F">Yu-Chiang Frank Wang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.05945v1-abstract-short" style="display: inline;"> Construction of a general-purpose post-recognition error corrector poses a crucial question: how can we most effectively train a model on a large mixture of domain datasets? The answer would lie in learning dataset-specific features and digesting their knowledge in a single model. Previous methods achieve this by having separate correction language models, resulting in a significant increase in pa… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.05945v1-abstract-full').style.display = 'inline'; document.getElementById('2411.05945v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.05945v1-abstract-full" style="display: none;"> Construction of a general-purpose post-recognition error corrector poses a crucial question: how can we most effectively train a model on a large mixture of domain datasets? The answer would lie in learning dataset-specific features and digesting their knowledge in a single model. Previous methods achieve this by having separate correction language models, resulting in a significant increase in parameters. In this work, we present Mixture-of-Experts as a solution, highlighting that MoEs are much more than a scalability tool. We propose a Multi-Task Correction MoE, where we train the experts to become an ``expert'' of speech-to-text, language-to-text and vision-to-text datasets by learning to route each dataset's tokens to its mapped expert. Experiments on the Open ASR Leaderboard show that we explore a new state-of-the-art performance by achieving an average relative $5.0$% WER reduction and substantial improvements in BLEU scores for speech and translation tasks. On zero-shot evaluation, NeKo outperforms GPT-3.5 and Claude-Opus with $15.5$% to $27.6$% relative WER reduction in the Hyporadise benchmark. NeKo performs competitively on grammar and post-OCR correction as a multi-task model. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.05945v1-abstract-full').style.display = 'none'; document.getElementById('2411.05945v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 8 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">NeKo work has been done in June 2024. NeKo LMs will be open source on https://huggingface.co/nvidia under the MIT license</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.05824">arXiv:2411.05824</a> <span> [<a href="https://arxiv.org/pdf/2411.05824">pdf</a>, <a href="https://arxiv.org/format/2411.05824">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Image and Video Processing">eess.IV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Navigating Distribution Shifts in Medical Image Analysis: A Survey </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Su%2C+Z">Zixian Su</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+J">Jingwei Guo</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xi Yang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Q">Qiufeng Wang</a>, <a href="/search/cs?searchtype=author&query=Coenen%2C+F">Frans Coenen</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+K">Kaizhu Huang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.05824v1-abstract-short" style="display: inline;"> Medical Image Analysis (MedIA) has become indispensable in modern healthcare, enhancing clinical diagnostics and personalized treatment. Despite the remarkable advancements supported by deep learning (DL) technologies, their practical deployment faces challenges due to distribution shifts, where models trained on specific datasets underperform across others from varying hospitals, regions, or pati… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.05824v1-abstract-full').style.display = 'inline'; document.getElementById('2411.05824v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.05824v1-abstract-full" style="display: none;"> Medical Image Analysis (MedIA) has become indispensable in modern healthcare, enhancing clinical diagnostics and personalized treatment. Despite the remarkable advancements supported by deep learning (DL) technologies, their practical deployment faces challenges due to distribution shifts, where models trained on specific datasets underperform across others from varying hospitals, regions, or patient populations. To navigate this issue, researchers have been actively developing strategies to increase the adaptability and robustness of DL models, enabling their effective use in unfamiliar and diverse environments. This paper systematically reviews approaches that apply DL techniques to MedIA systems affected by distribution shifts. Unlike traditional categorizations based on technical specifications, our approach is grounded in the real-world operational constraints faced by healthcare institutions. Specifically, we categorize the existing body of work into Joint Training, Federated Learning, Fine-tuning, and Domain Generalization, with each method tailored to distinct scenarios caused by Data Accessibility, Privacy Concerns, and Collaborative Protocols. This perspective equips researchers with a nuanced understanding of how DL can be strategically deployed to address distribution shifts in MedIA, ensuring diverse and robust medical applications. By delving deeper into these topics, we highlight potential pathways for future research that not only address existing limitations but also push the boundaries of deployable MedIA technologies. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.05824v1-abstract-full').style.display = 'none'; document.getElementById('2411.05824v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 5 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.05544">arXiv:2411.05544</a> <span> [<a href="https://arxiv.org/pdf/2411.05544">pdf</a>, <a href="https://arxiv.org/format/2411.05544">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Towards Lifelong Few-Shot Customization of Text-to-Image Diffusion </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Song%2C+N">Nan Song</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaofeng Yang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+Z">Ze Yang</a>, <a href="/search/cs?searchtype=author&query=Lin%2C+G">Guosheng Lin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.05544v1-abstract-short" style="display: inline;"> Lifelong few-shot customization for text-to-image diffusion aims to continually generalize existing models for new tasks with minimal data while preserving old knowledge. Current customization diffusion models excel in few-shot tasks but struggle with catastrophic forgetting problems in lifelong generations. In this study, we identify and categorize the catastrophic forgetting problems into two fo… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.05544v1-abstract-full').style.display = 'inline'; document.getElementById('2411.05544v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.05544v1-abstract-full" style="display: none;"> Lifelong few-shot customization for text-to-image diffusion aims to continually generalize existing models for new tasks with minimal data while preserving old knowledge. Current customization diffusion models excel in few-shot tasks but struggle with catastrophic forgetting problems in lifelong generations. In this study, we identify and categorize the catastrophic forgetting problems into two folds: relevant concepts forgetting and previous concepts forgetting. To address these challenges, we first devise a data-free knowledge distillation strategy to tackle relevant concepts forgetting. Unlike existing methods that rely on additional real data or offline replay of original concept data, our approach enables on-the-fly knowledge distillation to retain the previous concepts while learning new ones, without accessing any previous data. Second, we develop an In-Context Generation (ICGen) paradigm that allows the diffusion model to be conditioned upon the input vision context, which facilitates the few-shot generation and mitigates the issue of previous concepts forgetting. Extensive experiments show that the proposed Lifelong Few-Shot Diffusion (LFS-Diffusion) method can produce high-quality and accurate images while maintaining previously learned knowledge. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.05544v1-abstract-full').style.display = 'none'; document.getElementById('2411.05544v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 8 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.05311">arXiv:2411.05311</a> <span> [<a href="https://arxiv.org/pdf/2411.05311">pdf</a>, <a href="https://arxiv.org/format/2411.05311">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Robotics">cs.RO</span> </div> </div> <p class="title is-5 mathjax"> ZOPP: A Framework of Zero-shot Offboard Panoptic Perception for Autonomous Driving </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ma%2C+T">Tao Ma</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+H">Hongbin Zhou</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+Q">Qiusheng Huang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xuemeng Yang</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+J">Jianfei Guo</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+B">Bo Zhang</a>, <a href="/search/cs?searchtype=author&query=Dou%2C+M">Min Dou</a>, <a href="/search/cs?searchtype=author&query=Qiao%2C+Y">Yu Qiao</a>, <a href="/search/cs?searchtype=author&query=Shi%2C+B">Botian Shi</a>, <a href="/search/cs?searchtype=author&query=Li%2C+H">Hongsheng Li</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.05311v1-abstract-short" style="display: inline;"> Offboard perception aims to automatically generate high-quality 3D labels for autonomous driving (AD) scenes. Existing offboard methods focus on 3D object detection with closed-set taxonomy and fail to match human-level recognition capability on the rapidly evolving perception tasks. Due to heavy reliance on human labels and the prevalence of data imbalance and sparsity, a unified framework for of… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.05311v1-abstract-full').style.display = 'inline'; document.getElementById('2411.05311v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.05311v1-abstract-full" style="display: none;"> Offboard perception aims to automatically generate high-quality 3D labels for autonomous driving (AD) scenes. Existing offboard methods focus on 3D object detection with closed-set taxonomy and fail to match human-level recognition capability on the rapidly evolving perception tasks. Due to heavy reliance on human labels and the prevalence of data imbalance and sparsity, a unified framework for offboard auto-labeling various elements in AD scenes that meets the distinct needs of perception tasks is not being fully explored. In this paper, we propose a novel multi-modal Zero-shot Offboard Panoptic Perception (ZOPP) framework for autonomous driving scenes. ZOPP integrates the powerful zero-shot recognition capabilities of vision foundation models and 3D representations derived from point clouds. To the best of our knowledge, ZOPP represents a pioneering effort in the domain of multi-modal panoptic perception and auto labeling for autonomous driving scenes. We conduct comprehensive empirical studies and evaluations on Waymo open dataset to validate the proposed ZOPP on various perception tasks. To further explore the usability and extensibility of our proposed ZOPP, we also conduct experiments in downstream applications. The results further demonstrate the great potential of our ZOPP for real-world scenarios. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.05311v1-abstract-full').style.display = 'none'; document.getElementById('2411.05311v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 7 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted by NeurIPS 2024</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.04704">arXiv:2411.04704</a> <span> [<a href="https://arxiv.org/pdf/2411.04704">pdf</a>, <a href="https://arxiv.org/format/2411.04704">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Software Engineering">cs.SE</span> </div> </div> <p class="title is-5 mathjax"> Distinguishing LLM-generated from Human-written Code by Contrastive Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Xu%2C+X">Xiaodan Xu</a>, <a href="/search/cs?searchtype=author&query=Ni%2C+C">Chao Ni</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+X">Xinrong Guo</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+S">Shaoxuan Liu</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+X">Xiaoya Wang</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+K">Kui Liu</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaohu Yang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.04704v1-abstract-short" style="display: inline;"> Large language models (LLMs), such as ChatGPT released by OpenAI, have attracted significant attention from both industry and academia due to their demonstrated ability to generate high-quality content for various tasks. Despite the impressive capabilities of LLMs, there are growing concerns regarding their potential risks in various fields, such as news, education, and software engineering. Recen… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.04704v1-abstract-full').style.display = 'inline'; document.getElementById('2411.04704v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.04704v1-abstract-full" style="display: none;"> Large language models (LLMs), such as ChatGPT released by OpenAI, have attracted significant attention from both industry and academia due to their demonstrated ability to generate high-quality content for various tasks. Despite the impressive capabilities of LLMs, there are growing concerns regarding their potential risks in various fields, such as news, education, and software engineering. Recently, several commercial and open-source LLM-generated content detectors have been proposed, which, however, are primarily designed for detecting natural language content without considering the specific characteristics of program code. This paper aims to fill this gap by proposing a novel ChatGPT-generated code detector, CodeGPTSensor, based on a contrastive learning framework and a semantic encoder built with UniXcoder. To assess the effectiveness of CodeGPTSensor on differentiating ChatGPT-generated code from human-written code, we first curate a large-scale Human and Machine comparison Corpus (HMCorp), which includes 550K pairs of human-written and ChatGPT-generated code (i.e., 288K Python code pairs and 222K Java code pairs). Based on the HMCorp dataset, our qualitative and quantitative analysis of the characteristics of ChatGPT-generated code reveals the challenge and opportunity of distinguishing ChatGPT-generated code from human-written code with their representative features. Our experimental results indicate that CodeGPTSensor can effectively identify ChatGPT-generated code, outperforming all selected baselines. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.04704v1-abstract-full').style.display = 'none'; document.getElementById('2411.04704v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 7 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">30 pages, 6 figures, Accepted by TOSEM'24</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.03819">arXiv:2411.03819</a> <span> [<a href="https://arxiv.org/pdf/2411.03819">pdf</a>, <a href="https://arxiv.org/format/2411.03819">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> SA3DIP: Segment Any 3D Instance with Potential 3D Priors </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xi Yang</a>, <a href="/search/cs?searchtype=author&query=Gu%2C+X">Xu Gu</a>, <a href="/search/cs?searchtype=author&query=Yin%2C+X">Xingyilang Yin</a>, <a href="/search/cs?searchtype=author&query=Gao%2C+X">Xinbo Gao</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.03819v1-abstract-short" style="display: inline;"> The proliferation of 2D foundation models has sparked research into adapting them for open-world 3D instance segmentation. Recent methods introduce a paradigm that leverages superpoints as geometric primitives and incorporates 2D multi-view masks from Segment Anything model (SAM) as merging guidance, achieving outstanding zero-shot instance segmentation results. However, the limited use of 3D prio… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.03819v1-abstract-full').style.display = 'inline'; document.getElementById('2411.03819v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.03819v1-abstract-full" style="display: none;"> The proliferation of 2D foundation models has sparked research into adapting them for open-world 3D instance segmentation. Recent methods introduce a paradigm that leverages superpoints as geometric primitives and incorporates 2D multi-view masks from Segment Anything model (SAM) as merging guidance, achieving outstanding zero-shot instance segmentation results. However, the limited use of 3D priors restricts the segmentation performance. Previous methods calculate the 3D superpoints solely based on estimated normal from spatial coordinates, resulting in under-segmentation for instances with similar geometry. Besides, the heavy reliance on SAM and hand-crafted algorithms in 2D space suffers from over-segmentation due to SAM's inherent part-level segmentation tendency. To address these issues, we propose SA3DIP, a novel method for Segmenting Any 3D Instances via exploiting potential 3D Priors. Specifically, on one hand, we generate complementary 3D primitives based on both geometric and textural priors, which reduces the initial errors that accumulate in subsequent procedures. On the other hand, we introduce supplemental constraints from the 3D space by using a 3D detector to guide a further merging process. Furthermore, we notice a considerable portion of low-quality ground truth annotations in ScanNetV2 benchmark, which affect the fair evaluations. Thus, we present ScanNetV2-INS with complete ground truth labels and supplement additional instances for 3D class-agnostic instance segmentation. Experimental evaluations on various 2D-3D datasets demonstrate the effectiveness and robustness of our approach. Our code and proposed ScanNetV2-INS dataset are available HERE. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.03819v1-abstract-full').style.display = 'none'; document.getElementById('2411.03819v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 6 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.03059">arXiv:2411.03059</a> <span> [<a href="https://arxiv.org/pdf/2411.03059">pdf</a>, <a href="https://arxiv.org/format/2411.03059">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Enhancing DP-SGD through Non-monotonous Adaptive Scaling Gradient Weight </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Huang%2C+T">Tao Huang</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+Q">Qingyu Huang</a>, <a href="/search/cs?searchtype=author&query=Shi%2C+X">Xin Shi</a>, <a href="/search/cs?searchtype=author&query=Meng%2C+J">Jiayang Meng</a>, <a href="/search/cs?searchtype=author&query=Zheng%2C+G">Guolong Zheng</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xu Yang</a>, <a href="/search/cs?searchtype=author&query=Yi%2C+X">Xun Yi</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.03059v1-abstract-short" style="display: inline;"> In the domain of deep learning, the challenge of protecting sensitive data while maintaining model utility is significant. Traditional Differential Privacy (DP) techniques such as Differentially Private Stochastic Gradient Descent (DP-SGD) typically employ strategies like direct or per-sample adaptive gradient clipping. These methods, however, compromise model accuracy due to their critical influe… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.03059v1-abstract-full').style.display = 'inline'; document.getElementById('2411.03059v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.03059v1-abstract-full" style="display: none;"> In the domain of deep learning, the challenge of protecting sensitive data while maintaining model utility is significant. Traditional Differential Privacy (DP) techniques such as Differentially Private Stochastic Gradient Descent (DP-SGD) typically employ strategies like direct or per-sample adaptive gradient clipping. These methods, however, compromise model accuracy due to their critical influence on gradient handling, particularly neglecting the significant contribution of small gradients during later training stages. In this paper, we introduce an enhanced version of DP-SGD, named Differentially Private Per-sample Adaptive Scaling Clipping (DP-PSASC). This approach replaces traditional clipping with non-monotonous adaptive gradient scaling, which alleviates the need for intensive threshold setting and rectifies the disproportionate weighting of smaller gradients. Our contribution is twofold. First, we develop a novel gradient scaling technique that effectively assigns proper weights to gradients, particularly small ones, thus improving learning under differential privacy. Second, we integrate a momentum-based method into DP-PSASC to reduce bias from stochastic sampling, enhancing convergence rates. Our theoretical and empirical analyses confirm that DP-PSASC preserves privacy and delivers superior performance across diverse datasets, setting new standards for privacy-sensitive applications. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.03059v1-abstract-full').style.display = 'none'; document.getElementById('2411.03059v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 5 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.03053">arXiv:2411.03053</a> <span> [<a href="https://arxiv.org/pdf/2411.03053">pdf</a>, <a href="https://arxiv.org/format/2411.03053">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Gradient-Guided Conditional Diffusion Models for Private Image Reconstruction: Analyzing Adversarial Impacts of Differential Privacy and Denoising </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Huang%2C+T">Tao Huang</a>, <a href="/search/cs?searchtype=author&query=Meng%2C+J">Jiayang Meng</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+H">Hong Chen</a>, <a href="/search/cs?searchtype=author&query=Zheng%2C+G">Guolong Zheng</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xu Yang</a>, <a href="/search/cs?searchtype=author&query=Yi%2C+X">Xun Yi</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+H">Hua Wang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.03053v1-abstract-short" style="display: inline;"> We investigate the construction of gradient-guided conditional diffusion models for reconstructing private images, focusing on the adversarial interplay between differential privacy noise and the denoising capabilities of diffusion models. While current gradient-based reconstruction methods struggle with high-resolution images due to computational complexity and prior knowledge requirements, we pr… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.03053v1-abstract-full').style.display = 'inline'; document.getElementById('2411.03053v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.03053v1-abstract-full" style="display: none;"> We investigate the construction of gradient-guided conditional diffusion models for reconstructing private images, focusing on the adversarial interplay between differential privacy noise and the denoising capabilities of diffusion models. While current gradient-based reconstruction methods struggle with high-resolution images due to computational complexity and prior knowledge requirements, we propose two novel methods that require minimal modifications to the diffusion model's generation process and eliminate the need for prior knowledge. Our approach leverages the strong image generation capabilities of diffusion models to reconstruct private images starting from randomly generated noise, even when a small amount of differentially private noise has been added to the gradients. We also conduct a comprehensive theoretical analysis of the impact of differential privacy noise on the quality of reconstructed images, revealing the relationship among noise magnitude, the architecture of attacked models, and the attacker's reconstruction capability. Additionally, extensive experiments validate the effectiveness of our proposed methods and the accuracy of our theoretical findings, suggesting new directions for privacy risk auditing using conditional diffusion models. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.03053v1-abstract-full').style.display = 'none'; document.getElementById('2411.03053v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 5 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.02747">arXiv:2411.02747</a> <span> [<a href="https://arxiv.org/pdf/2411.02747">pdf</a>, <a href="https://arxiv.org/format/2411.02747">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Efficient Feature Aggregation and Scale-Aware Regression for Monocular 3D Object Detection </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yifan Wang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaochen Yang</a>, <a href="/search/cs?searchtype=author&query=Pu%2C+F">Fanqi Pu</a>, <a href="/search/cs?searchtype=author&query=Liao%2C+Q">Qingmin Liao</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+W">Wenming Yang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.02747v1-abstract-short" style="display: inline;"> Monocular 3D object detection has attracted great attention due to simplicity and low cost. Existing methods typically follow conventional 2D detection paradigms, first locating object centers and then predicting 3D attributes via neighboring features. However, these methods predominantly rely on progressive cross-scale feature aggregation and focus solely on local information, which may result in… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.02747v1-abstract-full').style.display = 'inline'; document.getElementById('2411.02747v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.02747v1-abstract-full" style="display: none;"> Monocular 3D object detection has attracted great attention due to simplicity and low cost. Existing methods typically follow conventional 2D detection paradigms, first locating object centers and then predicting 3D attributes via neighboring features. However, these methods predominantly rely on progressive cross-scale feature aggregation and focus solely on local information, which may result in a lack of global awareness and the omission of small-scale objects. In addition, due to large variation in object scales across different scenes and depths, inaccurate receptive fields often lead to background noise and degraded feature representation. To address these issues, we introduces MonoASRH, a novel monocular 3D detection framework composed of Efficient Hybrid Feature Aggregation Module (EH-FAM) and Adaptive Scale-Aware 3D Regression Head (ASRH). Specifically, EH-FAM employs multi-head attention with a global receptive field to extract semantic features for small-scale objects and leverages lightweight convolutional modules to efficiently aggregate visual features across different scales. The ASRH encodes 2D bounding box dimensions and then fuses scale features with the semantic features aggregated by EH-FAM through a scale-semantic feature fusion module. The scale-semantic feature fusion module guides ASRH in learning dynamic receptive field offsets, incorporating scale priors into 3D position prediction for better scale-awareness. Extensive experiments on the KITTI and Waymo datasets demonstrate that MonoASRH achieves state-of-the-art performance. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.02747v1-abstract-full').style.display = 'none'; document.getElementById('2411.02747v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 4 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.02430">arXiv:2411.02430</a> <span> [<a href="https://arxiv.org/pdf/2411.02430">pdf</a>, <a href="https://arxiv.org/format/2411.02430">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Generative Emotion Cause Explanation in Multimodal Conversations </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wang%2C+L">Lin Wang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaocui Yang</a>, <a href="/search/cs?searchtype=author&query=Feng%2C+S">Shi Feng</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+D">Daling Wang</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Y">Yifei Zhang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.02430v1-abstract-short" style="display: inline;"> Multimodal conversation, a crucial form of human communication, carries rich emotional content, making the exploration of the causes of emotions within it a research endeavor of significant importance. However, existing research on the causes of emotions typically uses clause selection methods to locate the reason utterance, without providing a detailed explanation of the emotional causes. In this… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.02430v1-abstract-full').style.display = 'inline'; document.getElementById('2411.02430v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.02430v1-abstract-full" style="display: none;"> Multimodal conversation, a crucial form of human communication, carries rich emotional content, making the exploration of the causes of emotions within it a research endeavor of significant importance. However, existing research on the causes of emotions typically uses clause selection methods to locate the reason utterance, without providing a detailed explanation of the emotional causes. In this paper, we propose a new task, \textbf{M}ultimodal \textbf{C}onversation \textbf{E}motion \textbf{C}ause \textbf{E}xplanation (MCECE), aiming to generate a detailed explanation of the emotional cause to the target utterance within a multimodal conversation scenario. Building upon the MELD dataset, we develop a new dataset (ECEM) that integrates video clips with detailed explanations of character emotions, facilitating an in-depth examination of the causal factors behind emotional expressions in multimodal conversations.A novel approach, FAME-Net, is further proposed, that harnesses the power of Large Language Models (LLMs) to analyze visual data and accurately interpret the emotions conveyed through facial expressions in videos. By exploiting the contagion effect of facial emotions, FAME-Net effectively captures the emotional causes of individuals engaged in conversations. Our experimental results on the newly constructed dataset show that FAME-Net significantly outperforms several excellent large language model baselines. Code and dataset are available at \url{https://github.com/3222345200/ECEMdataset.git} <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.02430v1-abstract-full').style.display = 'none'; document.getElementById('2411.02430v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 1 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.02337">arXiv:2411.02337</a> <span> [<a href="https://arxiv.org/pdf/2411.02337">pdf</a>, <a href="https://arxiv.org/format/2411.02337">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> </div> </div> <p class="title is-5 mathjax"> WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Qi%2C+Z">Zehan Qi</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+X">Xiao Liu</a>, <a href="/search/cs?searchtype=author&query=Iong%2C+I+L">Iat Long Iong</a>, <a href="/search/cs?searchtype=author&query=Lai%2C+H">Hanyu Lai</a>, <a href="/search/cs?searchtype=author&query=Sun%2C+X">Xueqiao Sun</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xinyue Yang</a>, <a href="/search/cs?searchtype=author&query=Sun%2C+J">Jiadai Sun</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+Y">Yu Yang</a>, <a href="/search/cs?searchtype=author&query=Yao%2C+S">Shuntian Yao</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+T">Tianjie Zhang</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+W">Wei Xu</a>, <a href="/search/cs?searchtype=author&query=Tang%2C+J">Jie Tang</a>, <a href="/search/cs?searchtype=author&query=Dong%2C+Y">Yuxiao Dong</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.02337v1-abstract-short" style="display: inline;"> Large language models (LLMs) have shown remarkable potential as autonomous agents, particularly in web-based tasks. However, existing LLM web agents heavily rely on expensive proprietary LLM APIs, while open LLMs lack the necessary decision-making capabilities. This paper introduces WebRL, a self-evolving online curriculum reinforcement learning framework designed to train high-performance web age… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.02337v1-abstract-full').style.display = 'inline'; document.getElementById('2411.02337v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.02337v1-abstract-full" style="display: none;"> Large language models (LLMs) have shown remarkable potential as autonomous agents, particularly in web-based tasks. However, existing LLM web agents heavily rely on expensive proprietary LLM APIs, while open LLMs lack the necessary decision-making capabilities. This paper introduces WebRL, a self-evolving online curriculum reinforcement learning framework designed to train high-performance web agents using open LLMs. WebRL addresses three key challenges in building LLM web agents, including the scarcity of training tasks, sparse feedback signals, and policy distribution drift in online learning. Specifically, WebRL incorporates 1) a self-evolving curriculum that generates new tasks from unsuccessful attempts, 2) a robust outcome-supervised reward model (ORM), and 3) adaptive reinforcement learning strategies to ensure consistent improvements. We apply WebRL to transform open Llama-3.1 and GLM-4 models into proficient web agents. On WebArena-Lite, WebRL improves the success rate of Llama-3.1-8B from 4.8% to 42.4%, and from 6.1% to 43% for GLM-4-9B. These open models significantly surpass the performance of GPT-4-Turbo (17.6%) and GPT-4o (13.9%) and outperform previous state-of-the-art web agents trained on open LLMs (AutoWebGLM, 18.2%). Our findings demonstrate WebRL's effectiveness in bridging the gap between open and proprietary LLM-based web agents, paving the way for more accessible and powerful autonomous web interaction systems. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.02337v1-abstract-full').style.display = 'none'; document.getElementById('2411.02337v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 4 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.02293">arXiv:2411.02293</a> <span> [<a href="https://arxiv.org/pdf/2411.02293">pdf</a>, <a href="https://arxiv.org/format/2411.02293">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Tencent Hunyuan3D-1.0: A Unified Framework for Text-to-3D and Image-to-3D Generation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xianghui Yang</a>, <a href="/search/cs?searchtype=author&query=Shi%2C+H">Huiwen Shi</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+B">Bowen Zhang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+F">Fan Yang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+J">Jiacheng Wang</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+H">Hongxu Zhao</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+X">Xinhai Liu</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+X">Xinzhou Wang</a>, <a href="/search/cs?searchtype=author&query=Lin%2C+Q">Qingxiang Lin</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+J">Jiaao Yu</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+L">Lifu Wang</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Z">Zhuo Chen</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+S">Sicong Liu</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Y">Yuhong Liu</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+Y">Yong Yang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+D">Di Wang</a>, <a href="/search/cs?searchtype=author&query=Jiang%2C+J">Jie Jiang</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+C">Chunchao Guo</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.02293v2-abstract-short" style="display: inline;"> While 3D generative models have greatly improved artists' workflows, the existing diffusion models for 3D generation suffer from slow generation and poor generalization. To address this issue, we propose a two-stage approach named Hunyuan3D-1.0 including a lite version and a standard version, that both support text- and image-conditioned generation. In the first stage, we employ a multi-view diffu… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.02293v2-abstract-full').style.display = 'inline'; document.getElementById('2411.02293v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.02293v2-abstract-full" style="display: none;"> While 3D generative models have greatly improved artists' workflows, the existing diffusion models for 3D generation suffer from slow generation and poor generalization. To address this issue, we propose a two-stage approach named Hunyuan3D-1.0 including a lite version and a standard version, that both support text- and image-conditioned generation. In the first stage, we employ a multi-view diffusion model that efficiently generates multi-view RGB in approximately 4 seconds. These multi-view images capture rich details of the 3D asset from different viewpoints, relaxing the tasks from single-view to multi-view reconstruction. In the second stage, we introduce a feed-forward reconstruction model that rapidly and faithfully reconstructs the 3D asset given the generated multi-view images in approximately 7 seconds. The reconstruction network learns to handle noises and in-consistency introduced by the multi-view diffusion and leverages the available information from the condition image to efficiently recover the 3D structure. Our framework involves the text-to-image model, i.e., Hunyuan-DiT, making it a unified framework to support both text- and image-conditioned 3D generation. Our standard version has 3x more parameters than our lite and other existing model. Our Hunyuan3D-1.0 achieves an impressive balance between speed and quality, significantly reducing generation time while maintaining the quality and diversity of the produced assets. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.02293v2-abstract-full').style.display = 'none'; document.getElementById('2411.02293v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 5 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 4 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Technical Report; 3D Generation</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.02057">arXiv:2411.02057</a> <span> [<a href="https://arxiv.org/pdf/2411.02057">pdf</a>, <a href="https://arxiv.org/format/2411.02057">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Exploiting Unlabeled Data with Multiple Expert Teachers for Open Vocabulary Aerial Object Detection and Its Orientation Adaptation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Li%2C+Y">Yan Li</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+W">Weiwei Guo</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xue Yang</a>, <a href="/search/cs?searchtype=author&query=Liao%2C+N">Ning Liao</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+S">Shaofeng Zhang</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+Y">Yi Yu</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+W">Wenxian Yu</a>, <a href="/search/cs?searchtype=author&query=Yan%2C+J">Junchi Yan</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.02057v1-abstract-short" style="display: inline;"> In recent years, aerial object detection has been increasingly pivotal in various earth observation applications. However, current algorithms are limited to detecting a set of pre-defined object categories, demanding sufficient annotated training samples, and fail to detect novel object categories. In this paper, we put forth a novel formulation of the aerial object detection problem, namely open-… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.02057v1-abstract-full').style.display = 'inline'; document.getElementById('2411.02057v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.02057v1-abstract-full" style="display: none;"> In recent years, aerial object detection has been increasingly pivotal in various earth observation applications. However, current algorithms are limited to detecting a set of pre-defined object categories, demanding sufficient annotated training samples, and fail to detect novel object categories. In this paper, we put forth a novel formulation of the aerial object detection problem, namely open-vocabulary aerial object detection (OVAD), which can detect objects beyond training categories without costly collecting new labeled data. We propose CastDet, a CLIP-activated student-teacher detection framework that serves as the first OVAD detector specifically designed for the challenging aerial scenario, where objects often exhibit weak appearance features and arbitrary orientations. Our framework integrates a robust localization teacher along with several box selection strategies to generate high-quality proposals for novel objects. Additionally, the RemoteCLIP model is adopted as an omniscient teacher, which provides rich knowledge to enhance classification capabilities for novel categories. A dynamic label queue is devised to maintain high-quality pseudo-labels during training. By doing so, the proposed CastDet boosts not only novel object proposals but also classification. Furthermore, we extend our approach from horizontal OVAD to oriented OVAD with tailored algorithm designs to effectively manage bounding box representation and pseudo-label generation. Extensive experiments for both tasks on multiple existing aerial object detection datasets demonstrate the effectiveness of our approach. The code is available at https://github.com/lizzy8587/CastDet. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.02057v1-abstract-full').style.display = 'none'; document.getElementById('2411.02057v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 4 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.01769">arXiv:2411.01769</a> <span> [<a href="https://arxiv.org/pdf/2411.01769">pdf</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> ARN-LSTM: A Multi-Stream Attention-Based Model for Action Recognition with Temporal Dynamics </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wang%2C+C">Chuanchuan Wang</a>, <a href="/search/cs?searchtype=author&query=Mohmamed%2C+A+S+A">Ahmad Sufril Azlan Mohmamed</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiao Yang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+X">Xiang Li</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.01769v1-abstract-short" style="display: inline;"> This paper presents ARN-LSTM, a novel multi-stream action recognition model designed to address the challenge of simultaneously capturing spatial motion and temporal dynamics in action sequences. Traditional methods often focus solely on spatial or temporal features, limiting their ability to comprehend complex human activities fully. Our proposed model integrates joint, motion, and temporal infor… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.01769v1-abstract-full').style.display = 'inline'; document.getElementById('2411.01769v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.01769v1-abstract-full" style="display: none;"> This paper presents ARN-LSTM, a novel multi-stream action recognition model designed to address the challenge of simultaneously capturing spatial motion and temporal dynamics in action sequences. Traditional methods often focus solely on spatial or temporal features, limiting their ability to comprehend complex human activities fully. Our proposed model integrates joint, motion, and temporal information through a multi-stream fusion architecture. Specifically, it comprises a joint stream for extracting skeleton features, a temporal stream for capturing dynamic temporal features, and an ARN-LSTM block that utilizes Time-Distributed Long Short-Term Memory (TD-LSTM) layers followed by an Attention Relation Network (ARN) to model temporal relations. The outputs from these streams are fused in a fully connected layer to provide the final action prediction. Evaluations on the NTU RGB+D 60 and NTU RGB+D 120 datasets demonstrate the effectiveness of our model, achieving effective performance, particularly in group activity recognition. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.01769v1-abstract-full').style.display = 'none'; document.getElementById('2411.01769v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 3 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.00927">arXiv:2411.00927</a> <span> [<a href="https://arxiv.org/pdf/2411.00927">pdf</a>, <a href="https://arxiv.org/format/2411.00927">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Human-Computer Interaction">cs.HC</span> </div> </div> <p class="title is-5 mathjax"> ReSpAct: Harmonizing Reasoning, Speaking, and Acting Towards Building Large Language Model-Based Conversational AI Agents </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Dongre%2C+V">Vardhan Dongre</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaocheng Yang</a>, <a href="/search/cs?searchtype=author&query=Acikgoz%2C+E+C">Emre Can Acikgoz</a>, <a href="/search/cs?searchtype=author&query=Dey%2C+S">Suvodip Dey</a>, <a href="/search/cs?searchtype=author&query=Tur%2C+G">Gokhan Tur</a>, <a href="/search/cs?searchtype=author&query=Hakkani-T%C3%BCr%2C+D">Dilek Hakkani-T眉r</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.00927v1-abstract-short" style="display: inline;"> Large language model (LLM)-based agents have been increasingly used to interact with external environments (e.g., games, APIs, etc.) and solve tasks. However, current frameworks do not enable these agents to work with users and interact with them to align on the details of their tasks and reach user-defined goals; instead, in ambiguous situations, these agents may make decisions based on assumptio… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.00927v1-abstract-full').style.display = 'inline'; document.getElementById('2411.00927v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.00927v1-abstract-full" style="display: none;"> Large language model (LLM)-based agents have been increasingly used to interact with external environments (e.g., games, APIs, etc.) and solve tasks. However, current frameworks do not enable these agents to work with users and interact with them to align on the details of their tasks and reach user-defined goals; instead, in ambiguous situations, these agents may make decisions based on assumptions. This work introduces ReSpAct (Reason, Speak, and Act), a novel framework that synergistically combines the essential skills for building task-oriented "conversational" agents. ReSpAct addresses this need for agents, expanding on the ReAct approach. The ReSpAct framework enables agents to interpret user instructions, reason about complex tasks, execute appropriate actions, and engage in dynamic dialogue to seek guidance, clarify ambiguities, understand user preferences, resolve problems, and use the intermediate feedback and responses of users to update their plans. We evaluated ReSpAct in environments supporting user interaction, such as task-oriented dialogue (MultiWOZ) and interactive decision-making (AlfWorld, WebShop). ReSpAct is flexible enough to incorporate dynamic user feedback and addresses prevalent issues like error propagation and agents getting stuck in reasoning loops. This results in more interpretable, human-like task-solving trajectories than relying solely on reasoning traces. In two interactive decision-making benchmarks, AlfWorld and WebShop, ReSpAct outperform the strong reasoning-only method ReAct by an absolute success rate of 6% and 4%, respectively. In the task-oriented dialogue benchmark MultiWOZ, ReSpAct improved Inform and Success scores by 5.5% and 3%, respectively. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.00927v1-abstract-full').style.display = 'none'; document.getElementById('2411.00927v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 1 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">30 pages, 9 Figures, 22 Tables</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.00882">arXiv:2411.00882</a> <span> [<a href="https://arxiv.org/pdf/2411.00882">pdf</a>, <a href="https://arxiv.org/format/2411.00882">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Technical Report for Soccernet 2023 -- Dense Video Captioning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ruan%2C+Z">Zheng Ruan</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+R">Ruixuan Liu</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+S">Shimin Chen</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+M">Mengying Zhou</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xinquan Yang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+W">Wei Li</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+C">Chen Chen</a>, <a href="/search/cs?searchtype=author&query=Shen%2C+W">Wei Shen</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.00882v1-abstract-short" style="display: inline;"> In the task of dense video captioning of Soccernet dataset, we propose to generate a video caption of each soccer action and locate the timestamp of the caption. Firstly, we apply Blip as our video caption framework to generate video captions. Then we locate the timestamp by using (1) multi-size sliding windows (2) temporal proposal generation and (3) proposal classification. </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.00882v1-abstract-full" style="display: none;"> In the task of dense video captioning of Soccernet dataset, we propose to generate a video caption of each soccer action and locate the timestamp of the caption. Firstly, we apply Blip as our video caption framework to generate video captions. Then we locate the timestamp by using (1) multi-size sliding windows (2) temporal proposal generation and (3) proposal classification. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.00882v1-abstract-full').style.display = 'none'; document.getElementById('2411.00882v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 31 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.00820">arXiv:2411.00820</a> <span> [<a href="https://arxiv.org/pdf/2411.00820">pdf</a>, <a href="https://arxiv.org/format/2411.00820">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Human-Computer Interaction">cs.HC</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> AutoGLM: Autonomous Foundation Agents for GUIs </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liu%2C+X">Xiao Liu</a>, <a href="/search/cs?searchtype=author&query=Qin%2C+B">Bo Qin</a>, <a href="/search/cs?searchtype=author&query=Liang%2C+D">Dongzhu Liang</a>, <a href="/search/cs?searchtype=author&query=Dong%2C+G">Guang Dong</a>, <a href="/search/cs?searchtype=author&query=Lai%2C+H">Hanyu Lai</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+H">Hanchen Zhang</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+H">Hanlin Zhao</a>, <a href="/search/cs?searchtype=author&query=Iong%2C+I+L">Iat Long Iong</a>, <a href="/search/cs?searchtype=author&query=Sun%2C+J">Jiadai Sun</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+J">Jiaqi Wang</a>, <a href="/search/cs?searchtype=author&query=Gao%2C+J">Junjie Gao</a>, <a href="/search/cs?searchtype=author&query=Shan%2C+J">Junjun Shan</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+K">Kangning Liu</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+S">Shudan Zhang</a>, <a href="/search/cs?searchtype=author&query=Yao%2C+S">Shuntian Yao</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+S">Siyi Cheng</a>, <a href="/search/cs?searchtype=author&query=Yao%2C+W">Wentao Yao</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+W">Wenyi Zhao</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+X">Xinghan Liu</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+X">Xinyi Liu</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+X">Xinying Chen</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xinyue Yang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+Y">Yang Yang</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+Y">Yifan Xu</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+Y">Yu Yang</a> , et al. (5 additional authors not shown) </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.00820v1-abstract-short" style="display: inline;"> We present AutoGLM, a new series in the ChatGLM family, designed to serve as foundation agents for autonomous control of digital devices through Graphical User Interfaces (GUIs). While foundation models excel at acquiring human knowledge, they often struggle with decision-making in dynamic real-world environments, limiting their progress toward artificial general intelligence. This limitation unde… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.00820v1-abstract-full').style.display = 'inline'; document.getElementById('2411.00820v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.00820v1-abstract-full" style="display: none;"> We present AutoGLM, a new series in the ChatGLM family, designed to serve as foundation agents for autonomous control of digital devices through Graphical User Interfaces (GUIs). While foundation models excel at acquiring human knowledge, they often struggle with decision-making in dynamic real-world environments, limiting their progress toward artificial general intelligence. This limitation underscores the importance of developing foundation agents capable of learning through autonomous environmental interactions by reinforcing existing models. Focusing on Web Browser and Phone as representative GUI scenarios, we have developed AutoGLM as a practical foundation agent system for real-world GUI interactions. Our approach integrates a comprehensive suite of techniques and infrastructures to create deployable agent systems suitable for user delivery. Through this development, we have derived two key insights: First, the design of an appropriate "intermediate interface" for GUI control is crucial, enabling the separation of planning and grounding behaviors, which require distinct optimization for flexibility and accuracy respectively. Second, we have developed a novel progressive training framework that enables self-evolving online curriculum reinforcement learning for AutoGLM. Our evaluations demonstrate AutoGLM's effectiveness across multiple domains. For web browsing, AutoGLM achieves a 55.2% success rate on VAB-WebArena-Lite (improving to 59.1% with a second attempt) and 96.2% on OpenTable evaluation tasks. In Android device control, AutoGLM attains a 36.2% success rate on AndroidLab (VAB-Mobile) and 89.7% on common tasks in popular Chinese APPs. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.00820v1-abstract-full').style.display = 'none'; document.getElementById('2411.00820v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 28 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.00787">arXiv:2411.00787</a> <span> [<a href="https://arxiv.org/pdf/2411.00787">pdf</a>, <a href="https://arxiv.org/format/2411.00787">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Networking and Internet Architecture">cs.NI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Optimization and Control">math.OC</span> </div> </div> <p class="title is-5 mathjax"> Novel operational algorithms for ride-pooling as on-demand feeder services </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Fan%2C+W">Wenbo Fan</a>, <a href="/search/cs?searchtype=author&query=Yan%2C+X">Xiaotian Yan</a>, <a href="/search/cs?searchtype=author&query=Sun%2C+Z">Zhanbo Sun</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xiaohui Yang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.00787v1-abstract-short" style="display: inline;"> Ride-pooling (RP) service, as a form of shared mobility, enables multiple riders with similar itineraries to share the same vehicle and split the fee. This makes RP a promising on-demand feeder service for patrons with a common trip end in urban transportation. We propose the RP as Feeder (RPaF) services with tailored operational algorithms. Specifically, we have developed (i) a batch-based matchi… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.00787v1-abstract-full').style.display = 'inline'; document.getElementById('2411.00787v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.00787v1-abstract-full" style="display: none;"> Ride-pooling (RP) service, as a form of shared mobility, enables multiple riders with similar itineraries to share the same vehicle and split the fee. This makes RP a promising on-demand feeder service for patrons with a common trip end in urban transportation. We propose the RP as Feeder (RPaF) services with tailored operational algorithms. Specifically, we have developed (i) a batch-based matching algorithm that pools a batch of requests within an optimized buffer distance to each RP vehicle; (ii) a dispatching algorithm that adaptively dispatches vehicles to pick up the matched requests for certain occupancy target; and (iii) a repositioning algorithm that relocates vehicles to unmatched requests based on their level of urgency. An agent-based microscopic simulation platform is designed to execute these operational algorithms (via the Operator module), generate spatially distributed random requests (Patron module), and account for traffic conditions (Vehicle module) in street networks. Extensive numerical experiments are conducted to showcase the effectiveness of RPaF services across various demand scenarios in typical morning rush hours. We compare RFaF with two on-demand feeder counterparts proposed in previous studies: Ride-Sharing as Feeder (RSaF) and Flexible-Route Feeder-Bus Transit (Flex-FBT). Comparisons reveal that given the same fleet size, RPaF generally outperforms RSaF in higher service rates (i.e., the percentage of requests served over all requests) and Flex-FBT in shorter average trip times for patrons. Lastly, we illustrate the implementation of RPaF in a real-world case study of the uptown Manhattan network (USA) using actual taxi trip data. The results demonstrate that RPaF effectively balances the level of service (service rate and patrons' average trip time) with operational costs (fleet size). <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.00787v1-abstract-full').style.display = 'none'; document.getElementById('2411.00787v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 17 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.00775">arXiv:2411.00775</a> <span> [<a href="https://arxiv.org/pdf/2411.00775">pdf</a>, <a href="https://arxiv.org/ps/2411.00775">ps</a>, <a href="https://arxiv.org/format/2411.00775">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> Dimension-free Private Mean Estimation for Anisotropic Distributions </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Dagan%2C+Y">Yuval Dagan</a>, <a href="/search/cs?searchtype=author&query=Jordan%2C+M+I">Michael I. Jordan</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+X">Xuelin Yang</a>, <a href="/search/cs?searchtype=author&query=Zakynthinou%2C+L">Lydia Zakynthinou</a>, <a href="/search/cs?searchtype=author&query=Zhivotovskiy%2C+N">Nikita Zhivotovskiy</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.00775v1-abstract-short" style="display: inline;"> We present differentially private algorithms for high-dimensional mean estimation. Previous private estimators on distributions over $\mathbb{R}^d$ suffer from a curse of dimensionality, as they require $惟(d^{1/2})$ samples to achieve non-trivial error, even in cases where $O(1)$ samples suffice without privacy. This rate is unavoidable when the distribution is isotropic, namely, when the covarian… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.00775v1-abstract-full').style.display = 'inline'; document.getElementById('2411.00775v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.00775v1-abstract-full" style="display: none;"> We present differentially private algorithms for high-dimensional mean estimation. Previous private estimators on distributions over $\mathbb{R}^d$ suffer from a curse of dimensionality, as they require $惟(d^{1/2})$ samples to achieve non-trivial error, even in cases where $O(1)$ samples suffice without privacy. This rate is unavoidable when the distribution is isotropic, namely, when the covariance is a multiple of the identity matrix, or when accuracy is measured with respect to the affine-invariant Mahalanobis distance. Yet, real-world data is often highly anisotropic, with signals concentrated on a small number of principal components. We develop estimators that are appropriate for such signals$\unicode{x2013}$our estimators are $(\varepsilon,未)$-differentially private and have sample complexity that is dimension-independent for anisotropic subgaussian distributions. Given $n$ samples from a distribution with known covariance-proxy $危$ and unknown mean $渭$, we present an estimator $\hat渭$ that achieves error $\|\hat渭-渭\|_2\leq 伪$, as long as $n\gtrsim\mathrm{tr}(危)/伪^2+ \mathrm{tr}(危^{1/2})/(伪\varepsilon)$. In particular, when $\pmb蟽^2=(蟽_1^2, \ldots, 蟽_d^2)$ are the singular values of $危$, we have $\mathrm{tr}(危)=\|\pmb蟽\|_2^2$ and $\mathrm{tr}(危^{1/2})=\|\pmb蟽\|_1$, and hence our bound avoids dimension-dependence when the signal is concentrated in a few principal components. We show that this is the optimal sample complexity for this task up to logarithmic factors. Moreover, for the case of unknown covariance, we present an algorithm whose sample complexity has improved dependence on the dimension, from $d^{1/2}$ to $d^{1/4}$. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.00775v1-abstract-full').style.display = 'none'; document.getElementById('2411.00775v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 1 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> </ol> <nav class="pagination is-small is-centered breathe-horizontal" role="navigation" aria-label="pagination"> <a href="" class="pagination-previous is-invisible">Previous </a> <a href="/search/?searchtype=author&query=Yang%2C+X&start=50" class="pagination-next" >Next </a> <ul class="pagination-list"> <li> <a href="/search/?searchtype=author&query=Yang%2C+X&start=0" class="pagination-link is-current" aria-label="Goto page 1">1 </a> </li> <li> <a href="/search/?searchtype=author&query=Yang%2C+X&start=50" class="pagination-link " aria-label="Page 2" aria-current="page">2 </a> </li> <li> <a href="/search/?searchtype=author&query=Yang%2C+X&start=100" class="pagination-link " aria-label="Page 3" aria-current="page">3 </a> </li> <li> <a href="/search/?searchtype=author&query=Yang%2C+X&start=150" class="pagination-link " aria-label="Page 4" aria-current="page">4 </a> </li> <li> <a href="/search/?searchtype=author&query=Yang%2C+X&start=200" class="pagination-link " aria-label="Page 5" aria-current="page">5 </a> </li> <li><span class="pagination-ellipsis">…</span></li> </ul> </nav> <div class="is-hidden-tablet">  <span class="help" style="display: inline-block;"><a href="https://github.com/arXiv/arxiv-search/releases">Search v0.5.6 released 2020-02-24</a>  </span> </div> </div> </main> <footer> <div class="columns is-desktop" role="navigation" aria-label="Secondary">  <div class="column"> <div class="columns"> <div class="column"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/about">About</a></li> <li><a href="https://info.arxiv.org/help">Help</a></li> </ul> </div> <div class="column"> <ul class="nav-spaced"> <li> <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><title>contact arXiv</title><desc>Click here to contact arXiv</desc><path d="M502.3 190.8c3.9-3.1 9.7-.2 9.7 4.7V400c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V195.6c0-5 5.7-7.8 9.7-4.7 22.4 17.4 52.1 39.5 154.1 113.6 21.1 15.4 56.7 47.8 92.2 47.6 35.7.3 72-32.8 92.3-47.6 102-74.1 131.6-96.3 154-113.7zM256 320c23.2.4 56.6-29.2 73.4-41.4 132.7-96.3 142.8-104.7 173.4-128.7 5.8-4.5 9.2-11.5 9.2-18.9v-19c0-26.5-21.5-48-48-48H48C21.5 64 0 85.5 0 112v19c0 7.4 3.4 14.3 9.2 18.9 30.6 23.9 40.7 32.4 173.4 128.7 16.8 12.2 50.2 41.8 73.4 41.4z"/></svg> <a href="https://info.arxiv.org/help/contact.html"> Contact</a> </li> <li> <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><title>subscribe to arXiv mailings</title><desc>Click here to subscribe</desc><path d="M476 3.2L12.5 270.6c-18.1 10.4-15.8 35.6 2.2 43.2L121 358.4l287.3-253.2c5.5-4.9 13.3 2.6 8.6 8.3L176 407v80.5c0 23.6 28.5 32.9 42.5 15.8L282 426l124.6 52.2c14.2 6 30.4-2.9 33-18.2l72-432C515 7.8 493.3-6.8 476 3.2z"/></svg> <a href="https://info.arxiv.org/help/subscribe"> Subscribe</a> </li> </ul> </div> </div> </div>   <div class="column"> <div class="columns"> <div class="column"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/help/license/index.html">Copyright</a></li> <li><a href="https://info.arxiv.org/help/policies/privacy_policy.html">Privacy Policy</a></li> </ul> </div> <div class="column sorry-app-links"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/help/web_accessibility.html">Web Accessibility Assistance</a></li> <li> <p class="help"> <a class="a11y-main-link" href="https://status.arxiv.org" target="_blank">arXiv Operational Status <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 256 512" class="icon filter-dark_grey" role="presentation"><path d="M224.3 273l-136 136c-9.4 9.4-24.6 9.4-33.9 0l-22.6-22.6c-9.4-9.4-9.4-24.6 0-33.9l96.4-96.4-96.4-96.4c-9.4-9.4-9.4-24.6 0-33.9L54.3 103c9.4-9.4 24.6-9.4 33.9 0l136 136c9.5 9.4 9.5 24.6.1 34z"/></svg></a><br> Get status notifications via <a class="is-link" href="https://subscribe.sorryapp.com/24846f03/email/new" target="_blank"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><path d="M502.3 190.8c3.9-3.1 9.7-.2 9.7 4.7V400c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V195.6c0-5 5.7-7.8 9.7-4.7 22.4 17.4 52.1 39.5 154.1 113.6 21.1 15.4 56.7 47.8 92.2 47.6 35.7.3 72-32.8 92.3-47.6 102-74.1 131.6-96.3 154-113.7zM256 320c23.2.4 56.6-29.2 73.4-41.4 132.7-96.3 142.8-104.7 173.4-128.7 5.8-4.5 9.2-11.5 9.2-18.9v-19c0-26.5-21.5-48-48-48H48C21.5 64 0 85.5 0 112v19c0 7.4 3.4 14.3 9.2 18.9 30.6 23.9 40.7 32.4 173.4 128.7 16.8 12.2 50.2 41.8 73.4 41.4z"/></svg>email</a> or <a class="is-link" href="https://subscribe.sorryapp.com/24846f03/slack/new" target="_blank"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 448 512" class="icon filter-black" role="presentation"><path d="M94.12 315.1c0 25.9-21.16 47.06-47.06 47.06S0 341 0 315.1c0-25.9 21.16-47.06 47.06-47.06h47.06v47.06zm23.72 0c0-25.9 21.16-47.06 47.06-47.06s47.06 21.16 47.06 47.06v117.84c0 25.9-21.16 47.06-47.06 47.06s-47.06-21.16-47.06-47.06V315.1zm47.06-188.98c-25.9 0-47.06-21.16-47.06-47.06S139 32 164.9 32s47.06 21.16 47.06 47.06v47.06H164.9zm0 23.72c25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06H47.06C21.16 243.96 0 222.8 0 196.9s21.16-47.06 47.06-47.06H164.9zm188.98 47.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06h-47.06V196.9zm-23.72 0c0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06V79.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06V196.9zM283.1 385.88c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06v-47.06h47.06zm0-23.72c-25.9 0-47.06-21.16-47.06-47.06 0-25.9 21.16-47.06 47.06-47.06h117.84c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06H283.1z"/></svg>slack</a> </p> </li> </ul> </div> </div> </div>  </div> </footer> <script src="https://static.arxiv.org/static/base/1.0.0a5/js/member_acknowledgement.js"></script> </body> </html>

CINXE.COM

Search | arXiv e-print repository