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 1–47 of 47 results for author: <span class="mathjax">Pang, Z</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=Pang%2C+Z">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="Pang, Z"> </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=Pang%2C+Z&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="Pang, Z"> <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> <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/2411.11406">arXiv:2411.11406</a> <span> [<a href="https://arxiv.org/pdf/2411.11406">pdf</a>, <a href="https://arxiv.org/format/2411.11406">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="Robotics">cs.RO</span> </div> </div> <p class="title is-5 mathjax"> Bridging the Resource Gap: Deploying Advanced Imitation Learning Models onto Affordable Embedded Platforms </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ge%2C+H">Haizhou Ge</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+R">Ruixiang Wang</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+Z">Zhu-ang Xu</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+H">Hongrui Zhu</a>, <a href="/search/cs?searchtype=author&query=Deng%2C+R">Ruichen Deng</a>, <a href="/search/cs?searchtype=author&query=Dong%2C+Y">Yuhang Dong</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zeyu Pang</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+G">Guyue Zhou</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+J">Junyu Zhang</a>, <a href="/search/cs?searchtype=author&query=Shi%2C+L">Lu Shi</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="2411.11406v1-abstract-short" style="display: inline;"> Advanced imitation learning with structures like the transformer is increasingly demonstrating its advantages in robotics. However, deploying these large-scale models on embedded platforms remains a major challenge. In this paper, we propose a pipeline that facilitates the migration of advanced imitation learning algorithms to edge devices. The process is achieved via an efficient model compressio… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11406v1-abstract-full').style.display = 'inline'; document.getElementById('2411.11406v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.11406v1-abstract-full" style="display: none;"> Advanced imitation learning with structures like the transformer is increasingly demonstrating its advantages in robotics. However, deploying these large-scale models on embedded platforms remains a major challenge. In this paper, we propose a pipeline that facilitates the migration of advanced imitation learning algorithms to edge devices. The process is achieved via an efficient model compression method and a practical asynchronous parallel method Temporal Ensemble with Dropped Actions (TEDA) that enhances the smoothness of operations. To show the efficiency of the proposed pipeline, large-scale imitation learning models are trained on a server and deployed on an edge device to complete various manipulation tasks. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.11406v1-abstract-full').style.display = 'none'; document.getElementById('2411.11406v1-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">Accepted by the 2024 IEEE International Conference on Robotics and Biomimetics (IEEE ROBIO 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/2410.07157">arXiv:2410.07157</a> <span> [<a href="https://arxiv.org/pdf/2410.07157">pdf</a>, <a href="https://arxiv.org/format/2410.07157">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="Computation and Language">cs.CL</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> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</span> </div> </div> <p class="title is-5 mathjax"> InstructG2I: Synthesizing Images from Multimodal Attributed Graphs </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jin%2C+B">Bowen Jin</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+B">Bingjun Guo</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yu-Xiong Wang</a>, <a href="/search/cs?searchtype=author&query=You%2C+J">Jiaxuan You</a>, <a href="/search/cs?searchtype=author&query=Han%2C+J">Jiawei Han</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="2410.07157v1-abstract-short" style="display: inline;"> In this paper, we approach an overlooked yet critical task Graph2Image: generating images from multimodal attributed graphs (MMAGs). This task poses significant challenges due to the explosion in graph size, dependencies among graph entities, and the need for controllability in graph conditions. To address these challenges, we propose a graph context-conditioned diffusion model called InstructG2I.… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.07157v1-abstract-full').style.display = 'inline'; document.getElementById('2410.07157v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.07157v1-abstract-full" style="display: none;"> In this paper, we approach an overlooked yet critical task Graph2Image: generating images from multimodal attributed graphs (MMAGs). This task poses significant challenges due to the explosion in graph size, dependencies among graph entities, and the need for controllability in graph conditions. To address these challenges, we propose a graph context-conditioned diffusion model called InstructG2I. InstructG2I first exploits the graph structure and multimodal information to conduct informative neighbor sampling by combining personalized page rank and re-ranking based on vision-language features. Then, a Graph-QFormer encoder adaptively encodes the graph nodes into an auxiliary set of graph prompts to guide the denoising process of diffusion. Finally, we propose graph classifier-free guidance, enabling controllable generation by varying the strength of graph guidance and multiple connected edges to a node. Extensive experiments conducted on three datasets from different domains demonstrate the effectiveness and controllability of our approach. The code is available at https://github.com/PeterGriffinJin/InstructG2I. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.07157v1-abstract-full').style.display = 'none'; document.getElementById('2410.07157v1-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 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 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">16 pages</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> NeurIPs 2024 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2409.20002">arXiv:2409.20002</a> <span> [<a href="https://arxiv.org/pdf/2409.20002">pdf</a>, <a href="https://arxiv.org/format/2409.20002">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"> The Early Bird Catches the Leak: Unveiling Timing Side Channels in LLM Serving Systems </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Song%2C+L">Linke Song</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zixuan Pang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+W">Wenhao Wang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Z">Zihao Wang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+X">XiaoFeng Wang</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+H">Hongbo Chen</a>, <a href="/search/cs?searchtype=author&query=Song%2C+W">Wei Song</a>, <a href="/search/cs?searchtype=author&query=Jin%2C+Y">Yier Jin</a>, <a href="/search/cs?searchtype=author&query=Meng%2C+D">Dan Meng</a>, <a href="/search/cs?searchtype=author&query=Hou%2C+R">Rui Hou</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="2409.20002v2-abstract-short" style="display: inline;"> The wide deployment of Large Language Models (LLMs) has given rise to strong demands for optimizing their inference performance. Today's techniques serving this purpose primarily focus on reducing latency and improving throughput through algorithmic and hardware enhancements, while largely overlooking their privacy side effects, particularly in a multi-user environment. In our research, for the fi… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.20002v2-abstract-full').style.display = 'inline'; document.getElementById('2409.20002v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2409.20002v2-abstract-full" style="display: none;"> The wide deployment of Large Language Models (LLMs) has given rise to strong demands for optimizing their inference performance. Today's techniques serving this purpose primarily focus on reducing latency and improving throughput through algorithmic and hardware enhancements, while largely overlooking their privacy side effects, particularly in a multi-user environment. In our research, for the first time, we discovered a set of new timing side channels in LLM systems, arising from shared caches and GPU memory allocations, which can be exploited to infer both confidential system prompts and those issued by other users. These vulnerabilities echo security challenges observed in traditional computing systems, highlighting an urgent need to address potential information leakage in LLM serving infrastructures. In this paper, we report novel attack strategies designed to exploit such timing side channels inherent in LLM deployments, specifically targeting the Key-Value (KV) cache and semantic cache widely used to enhance LLM inference performance. Our approach leverages timing measurements and classification models to detect cache hits, allowing an adversary to infer private prompts with high accuracy. We also propose a token-by-token search algorithm to efficiently recover shared prompt prefixes in the caches, showing the feasibility of stealing system prompts and those produced by peer users. Our experimental studies on black-box testing of popular online LLM services demonstrate that such privacy risks are completely realistic, with significant consequences. Our findings underscore the need for robust mitigation to protect LLM systems against such emerging threats. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.20002v2-abstract-full').style.display = 'none'; document.getElementById('2409.20002v2-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">v1</span> submitted 30 September, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 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">This work was submitted for review on Sept. 5, 2024, and the initial version was uploaded to Arxiv on Sept. 30, 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/2408.09919">arXiv:2408.09919</a> <span> [<a href="https://arxiv.org/pdf/2408.09919">pdf</a>, <a href="https://arxiv.org/format/2408.09919">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"> Long-Tail Temporal Action Segmentation with Group-wise Temporal Logit Adjustment </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhanzhong Pang</a>, <a href="/search/cs?searchtype=author&query=Sener%2C+F">Fadime Sener</a>, <a href="/search/cs?searchtype=author&query=Ramasubramanian%2C+S">Shrinivas Ramasubramanian</a>, <a href="/search/cs?searchtype=author&query=Yao%2C+A">Angela Yao</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="2408.09919v1-abstract-short" style="display: inline;"> Procedural activity videos often exhibit a long-tailed action distribution due to varying action frequencies and durations. However, state-of-the-art temporal action segmentation methods overlook the long tail and fail to recognize tail actions. Existing long-tail methods make class-independent assumptions and struggle to identify tail classes when applied to temporal segmentation frameworks. This… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.09919v1-abstract-full').style.display = 'inline'; document.getElementById('2408.09919v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2408.09919v1-abstract-full" style="display: none;"> Procedural activity videos often exhibit a long-tailed action distribution due to varying action frequencies and durations. However, state-of-the-art temporal action segmentation methods overlook the long tail and fail to recognize tail actions. Existing long-tail methods make class-independent assumptions and struggle to identify tail classes when applied to temporal segmentation frameworks. This work proposes a novel group-wise temporal logit adjustment~(G-TLA) framework that combines a group-wise softmax formulation while leveraging activity information and action ordering for logit adjustment. The proposed framework significantly improves in segmenting tail actions without any performance loss on head actions. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.09919v1-abstract-full').style.display = 'none'; document.getElementById('2408.09919v1-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 August, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 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 ECCV 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/2407.06089">arXiv:2407.06089</a> <span> [<a href="https://arxiv.org/pdf/2407.06089">pdf</a>, <a href="https://arxiv.org/format/2407.06089">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"> Merge, Ensemble, and Cooperate! A Survey on Collaborative Strategies in the Era of Large Language Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lu%2C+J">Jinliang Lu</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziliang Pang</a>, <a href="/search/cs?searchtype=author&query=Xiao%2C+M">Min Xiao</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+Y">Yaochen Zhu</a>, <a href="/search/cs?searchtype=author&query=Xia%2C+R">Rui Xia</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+J">Jiajun 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="2407.06089v1-abstract-short" style="display: inline;"> The remarkable success of Large Language Models (LLMs) has ushered natural language processing (NLP) research into a new era. Despite their diverse capabilities, LLMs trained on different corpora exhibit varying strengths and weaknesses, leading to challenges in maximizing their overall efficiency and versatility. To address these challenges, recent studies have explored collaborative strategies f… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.06089v1-abstract-full').style.display = 'inline'; document.getElementById('2407.06089v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2407.06089v1-abstract-full" style="display: none;"> The remarkable success of Large Language Models (LLMs) has ushered natural language processing (NLP) research into a new era. Despite their diverse capabilities, LLMs trained on different corpora exhibit varying strengths and weaknesses, leading to challenges in maximizing their overall efficiency and versatility. To address these challenges, recent studies have explored collaborative strategies for LLMs. This paper provides a comprehensive overview of this emerging research area, highlighting the motivation behind such collaborations. Specifically, we categorize collaborative strategies into three primary approaches: Merging, Ensemble, and Cooperation. Merging involves integrating multiple LLMs in the parameter space. Ensemble combines the outputs of various LLMs. Cooperation} leverages different LLMs to allow full play to their diverse capabilities for specific tasks. We provide in-depth introductions to these methods from different perspectives and discuss their potential applications. Additionally, we outline future research directions, hoping this work will catalyze further studies on LLM collaborations and paving the way for advanced NLP applications. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.06089v1-abstract-full').style.display = 'none'; document.getElementById('2407.06089v1-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 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2406.10100">arXiv:2406.10100</a> <span> [<a href="https://arxiv.org/pdf/2406.10100">pdf</a>, <a href="https://arxiv.org/format/2406.10100">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"> SkySenseGPT: A Fine-Grained Instruction Tuning Dataset and Model for Remote Sensing Vision-Language Understanding </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Luo%2C+J">Junwei Luo</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhen Pang</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Y">Yongjun Zhang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+T">Tingzhu Wang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+L">Linlin Wang</a>, <a href="/search/cs?searchtype=author&query=Dang%2C+B">Bo Dang</a>, <a href="/search/cs?searchtype=author&query=Lao%2C+J">Jiangwei Lao</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+J">Jian Wang</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+J">Jingdong Chen</a>, <a href="/search/cs?searchtype=author&query=Tan%2C+Y">Yihua Tan</a>, <a href="/search/cs?searchtype=author&query=Li%2C+Y">Yansheng 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="2406.10100v2-abstract-short" style="display: inline;"> Remote Sensing Large Multi-Modal Models (RSLMMs) are developing rapidly and showcase significant capabilities in remote sensing imagery (RSI) comprehension. However, due to the limitations of existing datasets, RSLMMs have shortcomings in understanding the rich semantic relations among objects in complex remote sensing scenes. To unlock RSLMMs' complex comprehension ability, we propose a large-sca… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.10100v2-abstract-full').style.display = 'inline'; document.getElementById('2406.10100v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2406.10100v2-abstract-full" style="display: none;"> Remote Sensing Large Multi-Modal Models (RSLMMs) are developing rapidly and showcase significant capabilities in remote sensing imagery (RSI) comprehension. However, due to the limitations of existing datasets, RSLMMs have shortcomings in understanding the rich semantic relations among objects in complex remote sensing scenes. To unlock RSLMMs' complex comprehension ability, we propose a large-scale instruction tuning dataset FIT-RS, containing 1,800,851 instruction samples. FIT-RS covers common interpretation tasks and innovatively introduces several complex comprehension tasks of escalating difficulty, ranging from relation reasoning to image-level scene graph generation. Based on FIT-RS, we build the FIT-RSFG benchmark. Furthermore, we establish a new benchmark to evaluate the fine-grained relation comprehension capabilities of LMMs, named FIT-RSRC. Based on combined instruction data, we propose SkySenseGPT, which achieves outstanding performance on both public datasets and FIT-RSFG, surpassing existing RSLMMs. We hope the FIT-RS dataset can enhance the relation comprehension capability of RSLMMs and provide a large-scale fine-grained data source for the remote sensing community. The dataset will be available at https://github.com/Luo-Z13/SkySenseGPT <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.10100v2-abstract-full').style.display = 'none'; document.getElementById('2406.10100v2-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 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 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, 5 figures, 19 tables, dataset and code see https://github.com/Luo-Z13/SkySenseGPT</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2406.08476">arXiv:2406.08476</a> <span> [<a href="https://arxiv.org/pdf/2406.08476">pdf</a>, <a href="https://arxiv.org/format/2406.08476">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"> RMem: Restricted Memory Banks Improve Video Object Segmentation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhou%2C+J">Junbao Zhou</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yu-Xiong 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="2406.08476v1-abstract-short" style="display: inline;"> With recent video object segmentation (VOS) benchmarks evolving to challenging scenarios, we revisit a simple but overlooked strategy: restricting the size of memory banks. This diverges from the prevalent practice of expanding memory banks to accommodate extensive historical information. Our specially designed "memory deciphering" study offers a pivotal insight underpinning such a strategy: expan… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.08476v1-abstract-full').style.display = 'inline'; document.getElementById('2406.08476v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2406.08476v1-abstract-full" style="display: none;"> With recent video object segmentation (VOS) benchmarks evolving to challenging scenarios, we revisit a simple but overlooked strategy: restricting the size of memory banks. This diverges from the prevalent practice of expanding memory banks to accommodate extensive historical information. Our specially designed "memory deciphering" study offers a pivotal insight underpinning such a strategy: expanding memory banks, while seemingly beneficial, actually increases the difficulty for VOS modules to decode relevant features due to the confusion from redundant information. By restricting memory banks to a limited number of essential frames, we achieve a notable improvement in VOS accuracy. This process balances the importance and freshness of frames to maintain an informative memory bank within a bounded capacity. Additionally, restricted memory banks reduce the training-inference discrepancy in memory lengths compared with continuous expansion. This fosters new opportunities in temporal reasoning and enables us to introduce the previously overlooked "temporal positional embedding." Finally, our insights are embodied in "RMem" ("R" for restricted), a simple yet effective VOS modification that excels at challenging VOS scenarios and establishes new state of the art for object state changes (on the VOST dataset) and long videos (on the Long Videos dataset). Our code and demo are available at https://restricted-memory.github.io/. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.08476v1-abstract-full').style.display = 'none'; document.getElementById('2406.08476v1-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 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 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">CVPR 2024, Project Page: https://restricted-memory.github.io/</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2403.09993">arXiv:2403.09993</a> <span> [<a href="https://arxiv.org/pdf/2403.09993">pdf</a>, <a href="https://arxiv.org/format/2403.09993">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="Image and Video Processing">eess.IV</span> </div> </div> <p class="title is-5 mathjax"> TRG-Net: An Interpretable and Controllable Rain Generator </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhiqiang Pang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+H">Hong Wang</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+Q">Qi Xie</a>, <a href="/search/cs?searchtype=author&query=Meng%2C+D">Deyu Meng</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+Z">Zongben 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="2403.09993v2-abstract-short" style="display: inline;"> Exploring and modeling rain generation mechanism is critical for augmenting paired data to ease training of rainy image processing models. Against this task, this study proposes a novel deep learning based rain generator, which fully takes the physical generation mechanism underlying rains into consideration and well encodes the learning of the fundamental rain factors (i.e., shape, orientation, l… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.09993v2-abstract-full').style.display = 'inline'; document.getElementById('2403.09993v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2403.09993v2-abstract-full" style="display: none;"> Exploring and modeling rain generation mechanism is critical for augmenting paired data to ease training of rainy image processing models. Against this task, this study proposes a novel deep learning based rain generator, which fully takes the physical generation mechanism underlying rains into consideration and well encodes the learning of the fundamental rain factors (i.e., shape, orientation, length, width and sparsity) explicitly into the deep network. Its significance lies in that the generator not only elaborately design essential elements of the rain to simulate expected rains, like conventional artificial strategies, but also finely adapt to complicated and diverse practical rainy images, like deep learning methods. By rationally adopting filter parameterization technique, we first time achieve a deep network that is finely controllable with respect to rain factors and able to learn the distribution of these factors purely from data. Our unpaired generation experiments demonstrate that the rain generated by the proposed rain generator is not only of higher quality, but also more effective for deraining and downstream tasks compared to current state-of-the-art rain generation methods. Besides, the paired data augmentation experiments, including both in-distribution and out-of-distribution (OOD), further validate the diversity of samples generated by our model for in-distribution deraining and OOD generalization tasks. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.09993v2-abstract-full').style.display = 'none'; document.getElementById('2403.09993v2-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> 29 April, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2402.17486">arXiv:2402.17486</a> <span> [<a href="https://arxiv.org/pdf/2402.17486">pdf</a>, <a href="https://arxiv.org/format/2402.17486">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"> MGE: A Training-Free and Efficient Model Generation and Enhancement Scheme </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wang%2C+X">Xuan Wang</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zeshan Pang</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+Y">Yuliang Lu</a>, <a href="/search/cs?searchtype=author&query=Yan%2C+X">Xuehu 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="2402.17486v1-abstract-short" style="display: inline;"> To provide a foundation for the research of deep learning models, the construction of model pool is an essential step. This paper proposes a Training-Free and Efficient Model Generation and Enhancement Scheme (MGE). This scheme primarily considers two aspects during the model generation process: the distribution of model parameters and model performance. Experiments result shows that generated mod… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.17486v1-abstract-full').style.display = 'inline'; document.getElementById('2402.17486v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2402.17486v1-abstract-full" style="display: none;"> To provide a foundation for the research of deep learning models, the construction of model pool is an essential step. This paper proposes a Training-Free and Efficient Model Generation and Enhancement Scheme (MGE). This scheme primarily considers two aspects during the model generation process: the distribution of model parameters and model performance. Experiments result shows that generated models are comparable to models obtained through normal training, and even superior in some cases. Moreover, the time consumed in generating models accounts for only 1\% of the time required for normal model training. More importantly, with the enhancement of Evolution-MGE, generated models exhibits competitive generalization ability in few-shot tasks. And the behavioral dissimilarity of generated models has the potential of adversarial defense. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.17486v1-abstract-full').style.display = 'none'; document.getElementById('2402.17486v1-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> 27 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2402.12770">arXiv:2402.12770</a> <span> [<a href="https://arxiv.org/pdf/2402.12770">pdf</a>, <a href="https://arxiv.org/format/2402.12770">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"> Acknowledgment of Emotional States: Generating Validating Responses for Empathetic Dialogue </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z+H">Zi Haur Pang</a>, <a href="/search/cs?searchtype=author&query=Fu%2C+Y">Yahui Fu</a>, <a href="/search/cs?searchtype=author&query=Lala%2C+D">Divesh Lala</a>, <a href="/search/cs?searchtype=author&query=Ochi%2C+K">Keiko Ochi</a>, <a href="/search/cs?searchtype=author&query=Inoue%2C+K">Koji Inoue</a>, <a href="/search/cs?searchtype=author&query=Kawahara%2C+T">Tatsuya Kawahara</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="2402.12770v1-abstract-short" style="display: inline;"> In the realm of human-AI dialogue, the facilitation of empathetic responses is important. Validation is one of the key communication techniques in psychology, which entails recognizing, understanding, and acknowledging others' emotional states, thoughts, and actions. This study introduces the first framework designed to engender empathetic dialogue with validating responses. Our approach incorpora… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.12770v1-abstract-full').style.display = 'inline'; document.getElementById('2402.12770v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2402.12770v1-abstract-full" style="display: none;"> In the realm of human-AI dialogue, the facilitation of empathetic responses is important. Validation is one of the key communication techniques in psychology, which entails recognizing, understanding, and acknowledging others' emotional states, thoughts, and actions. This study introduces the first framework designed to engender empathetic dialogue with validating responses. Our approach incorporates a tripartite module system: 1) validation timing detection, 2) users' emotional state identification, and 3) validating response generation. Utilizing Japanese EmpatheticDialogues dataset - a textual-based dialogue dataset consisting of 8 emotional categories from Plutchik's wheel of emotions - the Task Adaptive Pre-Training (TAPT) BERT-based model outperforms both random baseline and the ChatGPT performance, in term of F1-score, in all modules. Further validation of our model's efficacy is confirmed in its application to the TUT Emotional Storytelling Corpus (TESC), a speech-based dialogue dataset, by surpassing both random baseline and the ChatGPT. This consistent performance across both textual and speech-based dialogues underscores the effectiveness of our framework in fostering empathetic human-AI communication. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.12770v1-abstract-full').style.display = 'none'; document.getElementById('2402.12770v1-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 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 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">This paper has been accepted for presentation at International Workshop on Spoken Dialogue Systems Technology 2024 (IWSDS 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/2311.03194">arXiv:2311.03194</a> <span> [<a href="https://arxiv.org/pdf/2311.03194">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"> Few-shot Learning using Data Augmentation and Time-Frequency Transformation for Time Series Classification </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhang%2C+H">Hao Zhang</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhendong Pang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+J">Jiangpeng Wang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+T">Teng 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="2311.03194v1-abstract-short" style="display: inline;"> Deep neural networks (DNNs) that tackle the time series classification (TSC) task have provided a promising framework in signal processing. In real-world applications, as a data-driven model, DNNs are suffered from insufficient data. Few-shot learning has been studied to deal with this limitation. In this paper, we propose a novel few-shot learning framework through data augmentation, which involv… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.03194v1-abstract-full').style.display = 'inline'; document.getElementById('2311.03194v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2311.03194v1-abstract-full" style="display: none;"> Deep neural networks (DNNs) that tackle the time series classification (TSC) task have provided a promising framework in signal processing. In real-world applications, as a data-driven model, DNNs are suffered from insufficient data. Few-shot learning has been studied to deal with this limitation. In this paper, we propose a novel few-shot learning framework through data augmentation, which involves transformation through the time-frequency domain and the generation of synthetic images through random erasing. Additionally, we develop a sequence-spectrogram neural network (SSNN). This neural network model composes of two sub-networks: one utilizing 1D residual blocks to extract features from the input sequence while the other one employing 2D residual blocks to extract features from the spectrogram representation. In the experiments, comparison studies of different existing DNN models with/without data augmentation are conducted on an amyotrophic lateral sclerosis (ALS) dataset and a wind turbine fault (WTF) dataset. The experimental results manifest that our proposed method achieves 93.75% F1 score and 93.33% accuracy on the ALS datasets while 95.48% F1 score and 95.59% accuracy on the WTF datasets. Our methodology demonstrates its applicability of addressing the few-shot problems for time series classification. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.03194v1-abstract-full').style.display = 'none'; document.getElementById('2311.03194v1-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, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.12973">arXiv:2310.12973</a> <span> [<a href="https://arxiv.org/pdf/2310.12973">pdf</a>, <a href="https://arxiv.org/format/2310.12973">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> <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"> Frozen Transformers in Language Models Are Effective Visual Encoder Layers </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+Z">Ziyang Xie</a>, <a href="/search/cs?searchtype=author&query=Man%2C+Y">Yunze Man</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yu-Xiong 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="2310.12973v2-abstract-short" style="display: inline;"> This paper reveals that large language models (LLMs), despite being trained solely on textual data, are surprisingly strong encoders for purely visual tasks in the absence of language. Even more intriguingly, this can be achieved by a simple yet previously overlooked strategy -- employing a frozen transformer block from pre-trained LLMs as a constituent encoder layer to directly process visual tok… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.12973v2-abstract-full').style.display = 'inline'; document.getElementById('2310.12973v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.12973v2-abstract-full" style="display: none;"> This paper reveals that large language models (LLMs), despite being trained solely on textual data, are surprisingly strong encoders for purely visual tasks in the absence of language. Even more intriguingly, this can be achieved by a simple yet previously overlooked strategy -- employing a frozen transformer block from pre-trained LLMs as a constituent encoder layer to directly process visual tokens. Our work pushes the boundaries of leveraging LLMs for computer vision tasks, significantly departing from conventional practices that typically necessitate a multi-modal vision-language setup with associated language prompts, inputs, or outputs. We demonstrate that our approach consistently enhances performance across a diverse range of tasks, encompassing pure 2D and 3D visual recognition tasks (e.g., image and point cloud classification), temporal modeling tasks (e.g., action recognition), non-semantic tasks (e.g., motion forecasting), and multi-modal tasks (e.g., 2D/3D visual question answering and image-text retrieval). Such improvements are a general phenomenon, applicable to various types of LLMs (e.g., LLaMA and OPT) and different LLM transformer blocks. We additionally propose the information filtering hypothesis to explain the effectiveness of pre-trained LLMs in visual encoding -- the pre-trained LLM transformer blocks discern informative visual tokens and further amplify their effect. This hypothesis is empirically supported by the observation that the feature activation, after training with LLM transformer blocks, exhibits a stronger focus on relevant regions. We hope that our work inspires new perspectives on utilizing LLMs and deepening our understanding of their underlying mechanisms. Code is available at https://github.com/ziqipang/LM4VisualEncoding. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.12973v2-abstract-full').style.display = 'none'; document.getElementById('2310.12973v2-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 May, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 19 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </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">ICLR 2024 Spotlight. 23 pages, 13 figures. Code at https://github.com/ziqipang/LM4VisualEncoding</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.07405">arXiv:2310.07405</a> <span> [<a href="https://arxiv.org/pdf/2310.07405">pdf</a>, <a href="https://arxiv.org/ps/2310.07405">ps</a>, <a href="https://arxiv.org/format/2310.07405">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Signal Processing">eess.SP</span> </div> </div> <p class="title is-5 mathjax"> IRS Assisted Federated Learning A Broadband Over-the-Air Aggregation Approach </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhang%2C+D">Deyou Zhang</a>, <a href="/search/cs?searchtype=author&query=Xiao%2C+M">Ming Xiao</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhibo Pang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+L">Lihui Wang</a>, <a href="/search/cs?searchtype=author&query=Poor%2C+H+V">H. Vincent Poor</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="2310.07405v1-abstract-short" style="display: inline;"> We consider a broadband over-the-air computation empowered model aggregation approach for wireless federated learning (FL) systems and propose to leverage an intelligent reflecting surface (IRS) to combat wireless fading and noise. We first investigate the conventional node-selection based framework, where a few edge nodes are dropped in model aggregation to control the aggregation error. We analy… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.07405v1-abstract-full').style.display = 'inline'; document.getElementById('2310.07405v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.07405v1-abstract-full" style="display: none;"> We consider a broadband over-the-air computation empowered model aggregation approach for wireless federated learning (FL) systems and propose to leverage an intelligent reflecting surface (IRS) to combat wireless fading and noise. We first investigate the conventional node-selection based framework, where a few edge nodes are dropped in model aggregation to control the aggregation error. We analyze the performance of this node-selection based framework and derive an upper bound on its performance loss, which is shown to be related to the selected edge nodes. Then, we seek to minimize the mean-squared error (MSE) between the desired global gradient parameters and the actually received ones by optimizing the selected edge nodes, their transmit equalization coefficients, the IRS phase shifts, and the receive factors of the cloud server. By resorting to the matrix lifting technique and difference-of-convex programming, we successfully transform the formulated optimization problem into a convex one and solve it using off-the-shelf solvers. To improve learning performance, we further propose a weight-selection based FL framework. In such a framework, we assign each edge node a proper weight coefficient in model aggregation instead of discarding any of them to reduce the aggregation error, i.e., amplitude alignment of the received local gradient parameters from different edge nodes is not required. We also analyze the performance of this weight-selection based framework and derive an upper bound on its performance loss, followed by minimizing the MSE via optimizing the weight coefficients of the edge nodes, their transmit equalization coefficients, the IRS phase shifts, and the receive factors of the cloud server. Furthermore, we use the MNIST dataset for simulations to evaluate the performance of both node-selection and weight-selection based FL frameworks. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.07405v1-abstract-full').style.display = 'none'; document.getElementById('2310.07405v1-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 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </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">This paper has been accepted by IEEE Transactions on Wireless Communications</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.01351">arXiv:2310.01351</a> <span> [<a href="https://arxiv.org/pdf/2310.01351">pdf</a>, <a href="https://arxiv.org/format/2310.01351">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"> Streaming Motion Forecasting for Autonomous Driving </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Ramanan%2C+D">Deva Ramanan</a>, <a href="/search/cs?searchtype=author&query=Li%2C+M">Mengtian Li</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yu-Xiong 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="2310.01351v1-abstract-short" style="display: inline;"> Trajectory forecasting is a widely-studied problem for autonomous navigation. However, existing benchmarks evaluate forecasting based on independent snapshots of trajectories, which are not representative of real-world applications that operate on a continuous stream of data. To bridge this gap, we introduce a benchmark that continuously queries future trajectories on streaming data and we refer t… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.01351v1-abstract-full').style.display = 'inline'; document.getElementById('2310.01351v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.01351v1-abstract-full" style="display: none;"> Trajectory forecasting is a widely-studied problem for autonomous navigation. However, existing benchmarks evaluate forecasting based on independent snapshots of trajectories, which are not representative of real-world applications that operate on a continuous stream of data. To bridge this gap, we introduce a benchmark that continuously queries future trajectories on streaming data and we refer to it as "streaming forecasting." Our benchmark inherently captures the disappearance and re-appearance of agents, presenting the emergent challenge of forecasting for occluded agents, which is a safety-critical problem yet overlooked by snapshot-based benchmarks. Moreover, forecasting in the context of continuous timestamps naturally asks for temporal coherence between predictions from adjacent timestamps. Based on this benchmark, we further provide solutions and analysis for streaming forecasting. We propose a plug-and-play meta-algorithm called "Predictive Streamer" that can adapt any snapshot-based forecaster into a streaming forecaster. Our algorithm estimates the states of occluded agents by propagating their positions with multi-modal trajectories, and leverages differentiable filters to ensure temporal consistency. Both occlusion reasoning and temporal coherence strategies significantly improve forecasting quality, resulting in 25% smaller endpoint errors for occluded agents and 10-20% smaller fluctuations of trajectories. Our work is intended to generate interest within the community by highlighting the importance of addressing motion forecasting in its intrinsic streaming setting. Code is available at https://github.com/ziqipang/StreamingForecasting. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.01351v1-abstract-full').style.display = 'none'; document.getElementById('2310.01351v1-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> 2 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </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">IROS 2023, 8 pages, 9 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/2310.00033">arXiv:2310.00033</a> <span> [<a href="https://arxiv.org/pdf/2310.00033">pdf</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="Applied Physics">physics.app-ph</span> </div> </div> <p class="title is-5 mathjax"> OriWheelBot: An origami-wheeled robot </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liu%2C+J">Jie Liu</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zufeng Pang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+Z">Zhiyong Li</a>, <a href="/search/cs?searchtype=author&query=Wen%2C+G">Guilin Wen</a>, <a href="/search/cs?searchtype=author&query=Su%2C+Z">Zhoucheng Su</a>, <a href="/search/cs?searchtype=author&query=He%2C+J">Junfeng He</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+K">Kaiyue Liu</a>, <a href="/search/cs?searchtype=author&query=Jiang%2C+D">Dezheng Jiang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+Z">Zenan Li</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+S">Shouyan Chen</a>, <a href="/search/cs?searchtype=author&query=Tian%2C+Y">Yang Tian</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+Y+M">Yi Min Xie</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Z">Zhenpei Wang</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Z">Zhuangjian Liu</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="2310.00033v1-abstract-short" style="display: inline;"> Origami-inspired robots with multiple advantages, such as being lightweight, requiring less assembly, and exhibiting exceptional deformability, have received substantial and sustained attention. However, the existing origami-inspired robots are usually of limited functionalities and developing feature-rich robots is very challenging. Here, we report an origami-wheeled robot (OriWheelBot) with vari… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.00033v1-abstract-full').style.display = 'inline'; document.getElementById('2310.00033v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.00033v1-abstract-full" style="display: none;"> Origami-inspired robots with multiple advantages, such as being lightweight, requiring less assembly, and exhibiting exceptional deformability, have received substantial and sustained attention. However, the existing origami-inspired robots are usually of limited functionalities and developing feature-rich robots is very challenging. Here, we report an origami-wheeled robot (OriWheelBot) with variable width and outstanding sand walking versatility. The OriWheelBot's ability to adjust wheel width over obstacles is achieved by origami wheels made of Miura origami. An improved version, called iOriWheelBot, is also developed to automatically judge the width of the obstacles. Three actions, namely direct pass, variable width pass, and direct return, will be carried out depending on the width of the channel between the obstacles. We have identified two motion mechanisms, i.e., sand-digging and sand-pushing, with the latter being more conducive to walking on the sand. We have systematically examined numerous sand walking characteristics, including carrying loads, climbing a slope, walking on a slope, and navigating sand pits, small rocks, and sand traps. The OriWheelBot can change its width by 40%, has a loading-carrying ratio of 66.7% on flat sand and can climb a 17-degree sand incline. The OriWheelBot can be useful for planetary subsurface exploration and disaster area rescue. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.00033v1-abstract-full').style.display = 'none'; document.getElementById('2310.00033v1-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> 29 September, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </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">23 papes, 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/2307.15984">arXiv:2307.15984</a> <span> [<a href="https://arxiv.org/pdf/2307.15984">pdf</a>, <a href="https://arxiv.org/format/2307.15984">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Multimedia">cs.MM</span> </div> </div> <p class="title is-5 mathjax"> VATP360: Viewport Adaptive 360-Degree Video Streaming based on Tile Priority </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhiyu Pang</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="2307.15984v2-abstract-short" style="display: inline;"> 360-degree video becomes increasingly popular among users. In the current network bandwidth, serving high resolution 360 degree video to users is quite difficult. Most of the work has been devoted to the prediction of user viewports or tile-based adaptive algorithms. However, it is difficult to predict user viewports more accurately using only information such as user's historical viewports or vid… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.15984v2-abstract-full').style.display = 'inline'; document.getElementById('2307.15984v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2307.15984v2-abstract-full" style="display: none;"> 360-degree video becomes increasingly popular among users. In the current network bandwidth, serving high resolution 360 degree video to users is quite difficult. Most of the work has been devoted to the prediction of user viewports or tile-based adaptive algorithms. However, it is difficult to predict user viewports more accurately using only information such as user's historical viewports or video saliency maps. In this paper, we propose a viewport adaptive 360-degree video streaming method based on tile priority (VATP360), which tries to balance between the performance and the overhead. The proposed VATP360 consists of three main modules: viewport prediction, tile priority classification and bitrate allocation. In the viewport prediction module, object motion trajectory and predicted user's region-of-interest (ROI) are used to achieve accurate prediction of the user's future viewport. Then, the predicted viewport, along with the object motion trajectory, are fed into the proposed tile priority classification algorithm to assign different priorities to tiles, which would reduce the computational complexity of the bitrate allocation module. Finally in the bitrate allocation stage, we adaptively assign bitrates to tiles of different priority by reinforcement learning. Experimental results on publicly available datasets have demonstrated the effectiveness of the proposed method. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.15984v2-abstract-full').style.display = 'none'; document.getElementById('2307.15984v2-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> 27 August, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 29 July, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2306.11011">arXiv:2306.11011</a> <span> [<a href="https://arxiv.org/pdf/2306.11011">pdf</a>, <a href="https://arxiv.org/format/2306.11011">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"> virtCCA: Virtualized Arm Confidential Compute Architecture with TrustZone </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Xu%2C+X">Xiangyi Xu</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+W">Wenhao Wang</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+Y">Yongzheng Wu</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+C">Chenyu Wang</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+H">Huifeng Zhu</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+H">Haocheng Ma</a>, <a href="/search/cs?searchtype=author&query=Min%2C+Z">Zhennan Min</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zixuan Pang</a>, <a href="/search/cs?searchtype=author&query=Hou%2C+R">Rui Hou</a>, <a href="/search/cs?searchtype=author&query=Jin%2C+Y">Yier Jin</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="2306.11011v2-abstract-short" style="display: inline;"> ARM recently introduced the Confidential Compute Architecture (CCA) as part of the upcoming ARMv9-A architecture. CCA enables the support of confidential virtual machines (cVMs) within a separate world called the Realm world, providing protection from the untrusted normal world. While CCA offers a promising future for confidential computing, the widespread availability of CCA hardware is not expec… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.11011v2-abstract-full').style.display = 'inline'; document.getElementById('2306.11011v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2306.11011v2-abstract-full" style="display: none;"> ARM recently introduced the Confidential Compute Architecture (CCA) as part of the upcoming ARMv9-A architecture. CCA enables the support of confidential virtual machines (cVMs) within a separate world called the Realm world, providing protection from the untrusted normal world. While CCA offers a promising future for confidential computing, the widespread availability of CCA hardware is not expected in the near future, according to ARM's roadmap. To address this gap, we present virtCCA, an architecture that facilitates virtualized CCA using TrustZone, a mature hardware feature available on existing ARM platforms. Notably, virtCCA can be implemented on platforms equipped with the Secure EL2 (S-EL2) extension available from ARMv8.4 onwards, as well as on earlier platforms that lack S-EL2 support. virtCCA is fully compatible with the CCA specifications at the API level. We have developed the entire CCA software and firmware stack on top of virtCCA, including the enhancements to the normal world's KVM to support cVMs, and the TrustZone Management Monitor (TMM) that enforces isolation among cVMs and provides cVM life-cycle management. We have implemented virtCCA on real ARM servers, with and without S-EL2 support. Our evaluation, conducted on micro-benchmarks and macro-benchmarks, demonstrates that the overhead of running cVMs is acceptable compared to running normal-world VMs. Specifically, in a set of real-world workloads, the overhead of virtCCA-SEL2 is less than 29.5% for I/O intensive workloads, while virtCCA-EL3 outperforms the baseline in most cases. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.11011v2-abstract-full').style.display = 'none'; document.getElementById('2306.11011v2-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 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 19 June, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2305.08851">arXiv:2305.08851</a> <span> [<a href="https://arxiv.org/pdf/2305.08851">pdf</a>, <a href="https://arxiv.org/format/2305.08851">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"> MV-Map: Offboard HD-Map Generation with Multi-view Consistency </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Xie%2C+Z">Ziyang Xie</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yu-Xiong 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="2305.08851v3-abstract-short" style="display: inline;"> While bird's-eye-view (BEV) perception models can be useful for building high-definition maps (HD-Maps) with less human labor, their results are often unreliable and demonstrate noticeable inconsistencies in the predicted HD-Maps from different viewpoints. This is because BEV perception is typically set up in an 'onboard' manner, which restricts the computation and consequently prevents algorithms… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.08851v3-abstract-full').style.display = 'inline'; document.getElementById('2305.08851v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2305.08851v3-abstract-full" style="display: none;"> While bird's-eye-view (BEV) perception models can be useful for building high-definition maps (HD-Maps) with less human labor, their results are often unreliable and demonstrate noticeable inconsistencies in the predicted HD-Maps from different viewpoints. This is because BEV perception is typically set up in an 'onboard' manner, which restricts the computation and consequently prevents algorithms from reasoning multiple views simultaneously. This paper overcomes these limitations and advocates a more practical 'offboard' HD-Map generation setup that removes the computation constraints, based on the fact that HD-Maps are commonly reusable infrastructures built offline in data centers. To this end, we propose a novel offboard pipeline called MV-Map that capitalizes multi-view consistency and can handle an arbitrary number of frames with the key design of a 'region-centric' framework. In MV-Map, the target HD-Maps are created by aggregating all the frames of onboard predictions, weighted by the confidence scores assigned by an 'uncertainty network'. To further enhance multi-view consistency, we augment the uncertainty network with the global 3D structure optimized by a voxelized neural radiance field (Voxel-NeRF). Extensive experiments on nuScenes show that our MV-Map significantly improves the quality of HD-Maps, further highlighting the importance of offboard methods for HD-Map generation. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.08851v3-abstract-full').style.display = 'none'; document.getElementById('2305.08851v3-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 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 15 May, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2023. </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">ICCV 2023</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2302.03802">arXiv:2302.03802</a> <span> [<a href="https://arxiv.org/pdf/2302.03802">pdf</a>, <a href="https://arxiv.org/format/2302.03802">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> <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="Robotics">cs.RO</span> </div> </div> <p class="title is-5 mathjax"> Standing Between Past and Future: Spatio-Temporal Modeling for Multi-Camera 3D Multi-Object Tracking </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+J">Jie Li</a>, <a href="/search/cs?searchtype=author&query=Tokmakov%2C+P">Pavel Tokmakov</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+D">Dian Chen</a>, <a href="/search/cs?searchtype=author&query=Zagoruyko%2C+S">Sergey Zagoruyko</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yu-Xiong 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="2302.03802v2-abstract-short" style="display: inline;"> This work proposes an end-to-end multi-camera 3D multi-object tracking (MOT) framework. It emphasizes spatio-temporal continuity and integrates both past and future reasoning for tracked objects. Thus, we name it "Past-and-Future reasoning for Tracking" (PF-Track). Specifically, our method adapts the "tracking by attention" framework and represents tracked instances coherently over time with objec… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.03802v2-abstract-full').style.display = 'inline'; document.getElementById('2302.03802v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2302.03802v2-abstract-full" style="display: none;"> This work proposes an end-to-end multi-camera 3D multi-object tracking (MOT) framework. It emphasizes spatio-temporal continuity and integrates both past and future reasoning for tracked objects. Thus, we name it "Past-and-Future reasoning for Tracking" (PF-Track). Specifically, our method adapts the "tracking by attention" framework and represents tracked instances coherently over time with object queries. To explicitly use historical cues, our "Past Reasoning" module learns to refine the tracks and enhance the object features by cross-attending to queries from previous frames and other objects. The "Future Reasoning" module digests historical information and predicts robust future trajectories. In the case of long-term occlusions, our method maintains the object positions and enables re-association by integrating motion predictions. On the nuScenes dataset, our method improves AMOTA by a large margin and remarkably reduces ID-Switches by 90% compared to prior approaches, which is an order of magnitude less. The code and models are made available at https://github.com/TRI-ML/PF-Track. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.03802v2-abstract-full').style.display = 'none'; document.getElementById('2302.03802v2-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 April, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 7 February, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2023. </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">CVPR 2023 Camera Ready, 15 pages, 8 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/2212.00998">arXiv:2212.00998</a> <span> [<a href="https://arxiv.org/pdf/2212.00998">pdf</a>, <a href="https://arxiv.org/format/2212.00998">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"> Credit Assignment for Trained Neural Networks Based on Koopman Operator Theory </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liang%2C+Z">Zhen Liang</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+C">Changyuan Zhao</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+W">Wanwei Liu</a>, <a href="/search/cs?searchtype=author&query=Xue%2C+B">Bai Xue</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+W">Wenjing Yang</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhengbin Pang</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="2212.00998v1-abstract-short" style="display: inline;"> Credit assignment problem of neural networks refers to evaluating the credit of each network component to the final outputs. For an untrained neural network, approaches to tackling it have made great contributions to parameter update and model revolution during the training phase. This problem on trained neural networks receives rare attention, nevertheless, it plays an increasingly important role… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2212.00998v1-abstract-full').style.display = 'inline'; document.getElementById('2212.00998v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2212.00998v1-abstract-full" style="display: none;"> Credit assignment problem of neural networks refers to evaluating the credit of each network component to the final outputs. For an untrained neural network, approaches to tackling it have made great contributions to parameter update and model revolution during the training phase. This problem on trained neural networks receives rare attention, nevertheless, it plays an increasingly important role in neural network patch, specification and verification. Based on Koopman operator theory, this paper presents an alternative perspective of linear dynamics on dealing with the credit assignment problem for trained neural networks. Regarding a neural network as the composition of sub-dynamics series, we utilize step-delay embedding to capture snapshots of each component, characterizing the established mapping as exactly as possible. To circumvent the dimension-difference problem encountered during the embedding, a composition and decomposition of an auxiliary linear layer, termed minimal linear dimension alignment, is carefully designed with rigorous formal guarantee. Afterwards, each component is approximated by a Koopman operator and we derive the Jacobian matrix and its corresponding determinant, similar to backward propagation. Then, we can define a metric with algebraic interpretability for the credit assignment of each network component. Moreover, experiments conducted on typical neural networks demonstrate the effectiveness of the proposed method. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2212.00998v1-abstract-full').style.display = 'none'; document.getElementById('2212.00998v1-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> 2 December, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2022. </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, 4 figures</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 68T01 <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> I.2.0 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2211.10056">arXiv:2211.10056</a> <span> [<a href="https://arxiv.org/pdf/2211.10056">pdf</a>, <a href="https://arxiv.org/format/2211.10056">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"> Contrastive Losses Are Natural Criteria for Unsupervised Video Summarization </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zongshang Pang</a>, <a href="/search/cs?searchtype=author&query=Nakashima%2C+Y">Yuta Nakashima</a>, <a href="/search/cs?searchtype=author&query=Otani%2C+M">Mayu Otani</a>, <a href="/search/cs?searchtype=author&query=Nagahara%2C+H">Hajime Nagahara</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="2211.10056v1-abstract-short" style="display: inline;"> Video summarization aims to select the most informative subset of frames in a video to facilitate efficient video browsing. Unsupervised methods usually rely on heuristic training objectives such as diversity and representativeness. However, such methods need to bootstrap the online-generated summaries to compute the objectives for importance score regression. We consider such a pipeline inefficie… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2211.10056v1-abstract-full').style.display = 'inline'; document.getElementById('2211.10056v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2211.10056v1-abstract-full" style="display: none;"> Video summarization aims to select the most informative subset of frames in a video to facilitate efficient video browsing. Unsupervised methods usually rely on heuristic training objectives such as diversity and representativeness. However, such methods need to bootstrap the online-generated summaries to compute the objectives for importance score regression. We consider such a pipeline inefficient and seek to directly quantify the frame-level importance with the help of contrastive losses in the representation learning literature. Leveraging the contrastive losses, we propose three metrics featuring a desirable key frame: local dissimilarity, global consistency, and uniqueness. With features pre-trained on the image classification task, the metrics can already yield high-quality importance scores, demonstrating competitive or better performance than past heavily-trained methods. We show that by refining the pre-trained features with a lightweight contrastively learned projection module, the frame-level importance scores can be further improved, and the model can also leverage a large number of random videos and generalize to test videos with decent performance. Code available at https://github.com/pangzss/pytorch-CTVSUM. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2211.10056v1-abstract-full').style.display = 'none'; document.getElementById('2211.10056v1-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, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2022. </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">To appear in WACV2023</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2209.11553">arXiv:2209.11553</a> <span> [<a href="https://arxiv.org/pdf/2209.11553">pdf</a>, <a href="https://arxiv.org/format/2209.11553">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"> On Efficient Reinforcement Learning for Full-length Game of StarCraft II </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liu%2C+R">Ruo-Ze Liu</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhen-Jia Pang</a>, <a href="/search/cs?searchtype=author&query=Meng%2C+Z">Zhou-Yu Meng</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+W">Wenhai Wang</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+Y">Yang Yu</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+T">Tong 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="2209.11553v1-abstract-short" style="display: inline;"> StarCraft II (SC2) poses a grand challenge for reinforcement learning (RL), of which the main difficulties include huge state space, varying action space, and a long time horizon. In this work, we investigate a set of RL techniques for the full-length game of StarCraft II. We investigate a hierarchical RL approach involving extracted macro-actions and a hierarchical architecture of neural networks… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2209.11553v1-abstract-full').style.display = 'inline'; document.getElementById('2209.11553v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2209.11553v1-abstract-full" style="display: none;"> StarCraft II (SC2) poses a grand challenge for reinforcement learning (RL), of which the main difficulties include huge state space, varying action space, and a long time horizon. In this work, we investigate a set of RL techniques for the full-length game of StarCraft II. We investigate a hierarchical RL approach involving extracted macro-actions and a hierarchical architecture of neural networks. We investigate a curriculum transfer training procedure and train the agent on a single machine with 4 GPUs and 48 CPU threads. On a 64x64 map and using restrictive units, we achieve a win rate of 99% against the level-1 built-in AI. Through the curriculum transfer learning algorithm and a mixture of combat models, we achieve a 93% win rate against the most difficult non-cheating level built-in AI (level-7). In this extended version of the paper, we improve our architecture to train the agent against the cheating level AIs and achieve the win rate against the level-8, level-9, and level-10 AIs as 96%, 97%, and 94%, respectively. Our codes are at https://github.com/liuruoze/HierNet-SC2. To provide a baseline referring the AlphaStar for our work as well as the research and open-source community, we reproduce a scaled-down version of it, mini-AlphaStar (mAS). The latest version of mAS is 1.07, which can be trained on the raw action space which has 564 actions. It is designed to run training on a single common machine, by making the hyper-parameters adjustable. We then compare our work with mAS using the same resources and show that our method is more effective. The codes of mini-AlphaStar are at https://github.com/liuruoze/mini-AlphaStar. We hope our study could shed some light on the future research of efficient reinforcement learning on SC2 and other large-scale games. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2209.11553v1-abstract-full').style.display = 'none'; document.getElementById('2209.11553v1-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> 23 September, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2022. </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">48 pages,21 figures</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> JAIR, 75 (2022), 213-260 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2207.06718">arXiv:2207.06718</a> <span> [<a href="https://arxiv.org/pdf/2207.06718">pdf</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="Hardware Architecture">cs.AR</span> </div> <div class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1109/IECON49645.2022.9968471">10.1109/IECON49645.2022.9968471 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Hardware-in-the-Loop Simulation for Evaluating Communication Impacts on the Wireless-Network-Controlled Robots </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lv%2C+H">Honghao Lv</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhibo Pang</a>, <a href="/search/cs?searchtype=author&query=Xiao%2C+M">Ming Xiao</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+G">Geng 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="2207.06718v2-abstract-short" style="display: inline;"> More and more robot automation applications have changed to wireless communication, and network performance has a growing impact on robotic systems. This study proposes a hardware-in-the-loop (HiL) simulation methodology for connecting the simulated robot platform to real network devices. This project seeks to provide robotic engineers and researchers with the capability to experiment without heav… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2207.06718v2-abstract-full').style.display = 'inline'; document.getElementById('2207.06718v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2207.06718v2-abstract-full" style="display: none;"> More and more robot automation applications have changed to wireless communication, and network performance has a growing impact on robotic systems. This study proposes a hardware-in-the-loop (HiL) simulation methodology for connecting the simulated robot platform to real network devices. This project seeks to provide robotic engineers and researchers with the capability to experiment without heavily modifying the original controller and get more realistic test results that correlate with actual network conditions. We deployed this HiL simulation system in two common cases for wireless-network-controlled robotic applications: (1) safe multi-robot coordination for mobile robots, and (2) human-motion-based teleoperation for manipulators. The HiL simulation system is deployed and tested under various network conditions in all circumstances. The experiment results are analyzed and compared with the previous simulation methods, demonstrating that the proposed HiL simulation methodology can identify a more reliable communication impact on robot systems. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2207.06718v2-abstract-full').style.display = 'none'; document.getElementById('2207.06718v2-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 September, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 July, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2022. </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">6 pages, 11 figures, to appear in 48th Annual Conference of the Industrial Electronics Society IECON 2022 Conference</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2207.05267">arXiv:2207.05267</a> <span> [<a href="https://arxiv.org/pdf/2207.05267">pdf</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link 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> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Instrumentation and Detectors">physics.ins-det</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Optics">physics.optics</span> </div> <div class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1364/OE.470529">10.1364/OE.470529 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Indoor optical fiber eavesdropping approach and its avoidance </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hao%2C+H">Haiqing Hao</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhongwang Pang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+G">Guan Wang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+B">Bo 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="2207.05267v2-abstract-short" style="display: inline;"> The optical fiber network has become a worldwide infrastructure. In addition to the basic functions in telecommunication, its sensing ability has attracted more and more attention. In this paper, we discuss the risk of household fiber being used for eavesdropping and demonstrate its performance in the lab. Using a 3-meter tail fiber in front of the household optical modem, voices of normal human s… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2207.05267v2-abstract-full').style.display = 'inline'; document.getElementById('2207.05267v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2207.05267v2-abstract-full" style="display: none;"> The optical fiber network has become a worldwide infrastructure. In addition to the basic functions in telecommunication, its sensing ability has attracted more and more attention. In this paper, we discuss the risk of household fiber being used for eavesdropping and demonstrate its performance in the lab. Using a 3-meter tail fiber in front of the household optical modem, voices of normal human speech can be eavesdropped by a laser interferometer and recovered 1.1 km away. The detection distance limit and system noise are analyzed quantitatively. We also give some practical ways to prevent eavesdropping through household fiber. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2207.05267v2-abstract-full').style.display = 'none'; document.getElementById('2207.05267v2-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 August, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 11 July, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2022. </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">8 pages, 4 figures, submitted to Optics Express</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2203.00770">arXiv:2203.00770</a> <span> [<a href="https://arxiv.org/pdf/2203.00770">pdf</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Signal Processing">eess.SP</span> </div> </div> <p class="title is-5 mathjax"> Short-Packet Interleaver against Impulse Interference in Practical Industrial Environments </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhan%2C+M">Ming Zhan</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhibo Pang</a>, <a href="/search/cs?searchtype=author&query=Dzung%2C+D">Dacfey Dzung</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+K">Kan Yu</a>, <a href="/search/cs?searchtype=author&query=Xiao%2C+M">Ming Xiao</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="2203.00770v1-abstract-short" style="display: inline;"> The most common cause of transmission failure in Wireless High Performance (WirelessHP) target industry environments is impulse interference. As interleavers are commonly used to improve the reliability on the Orthogonal Frequency Division Multiplexing (OFDM) symbol level for long packet transmission, this paper considers the feasibility of applying short-packet bit interleaving to enhance the imp… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2203.00770v1-abstract-full').style.display = 'inline'; document.getElementById('2203.00770v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2203.00770v1-abstract-full" style="display: none;"> The most common cause of transmission failure in Wireless High Performance (WirelessHP) target industry environments is impulse interference. As interleavers are commonly used to improve the reliability on the Orthogonal Frequency Division Multiplexing (OFDM) symbol level for long packet transmission, this paper considers the feasibility of applying short-packet bit interleaving to enhance the impulse/burst interference resisting capability on both OFDM symbol and frame level. Using the Universal Software Radio Peripherals (USRP) and PC hardware platform, the Packet Error Rate (PER) performance of interleaved coded short-packet transmission with Convolutional Codes (CC), Reed-Solomon codes (RS) and RS+CC concatenated codes are tested and analyzed. Applying the IEEE 1613 standard for impulse interference generation, extensive PER tests of CC(1=2) and RS(31; 21)+CC(1=2) concatenated codes are performed. With practical experiments, we prove the effectiveness of bit in terleaved coded short-packet transmission in real factory environments. We also investigate how PER performance depends on the interleavers, codes and impulse interference power and frequency. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2203.00770v1-abstract-full').style.display = 'none'; document.getElementById('2203.00770v1-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 March, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2022. </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, 12 figures, submitted to IEEE Transactions on Wireless Communications</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2112.06375">arXiv:2112.06375</a> <span> [<a href="https://arxiv.org/pdf/2112.06375">pdf</a>, <a href="https://arxiv.org/format/2112.06375">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"> Embracing Single Stride 3D Object Detector with Sparse Transformer </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Fan%2C+L">Lue Fan</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+T">Tianyuan Zhang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yu-Xiong Wang</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+H">Hang Zhao</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+F">Feng Wang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+N">Naiyan Wang</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Z">Zhaoxiang 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="2112.06375v1-abstract-short" style="display: inline;"> In LiDAR-based 3D object detection for autonomous driving, the ratio of the object size to input scene size is significantly smaller compared to 2D detection cases. Overlooking this difference, many 3D detectors directly follow the common practice of 2D detectors, which downsample the feature maps even after quantizing the point clouds. In this paper, we start by rethinking how such multi-stride s… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2112.06375v1-abstract-full').style.display = 'inline'; document.getElementById('2112.06375v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2112.06375v1-abstract-full" style="display: none;"> In LiDAR-based 3D object detection for autonomous driving, the ratio of the object size to input scene size is significantly smaller compared to 2D detection cases. Overlooking this difference, many 3D detectors directly follow the common practice of 2D detectors, which downsample the feature maps even after quantizing the point clouds. In this paper, we start by rethinking how such multi-stride stereotype affects the LiDAR-based 3D object detectors. Our experiments point out that the downsampling operations bring few advantages, and lead to inevitable information loss. To remedy this issue, we propose Single-stride Sparse Transformer (SST) to maintain the original resolution from the beginning to the end of the network. Armed with transformers, our method addresses the problem of insufficient receptive field in single-stride architectures. It also cooperates well with the sparsity of point clouds and naturally avoids expensive computation. Eventually, our SST achieves state-of-the-art results on the large scale Waymo Open Dataset. It is worth mentioning that our method can achieve exciting performance (83.8 LEVEL 1 AP on validation split) on small object (pedestrian) detection due to the characteristic of single stride. Codes will be released at https://github.com/TuSimple/SST <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2112.06375v1-abstract-full').style.display = 'none'; document.getElementById('2112.06375v1-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 December, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2021. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2111.13672">arXiv:2111.13672</a> <span> [<a href="https://arxiv.org/pdf/2111.13672">pdf</a>, <a href="https://arxiv.org/format/2111.13672">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"> Immortal Tracker: Tracklet Never Dies </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wang%2C+Q">Qitai Wang</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Y">Yuntao Chen</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+N">Naiyan Wang</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Z">Zhaoxiang 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="2111.13672v1-abstract-short" style="display: inline;"> Previous online 3D Multi-Object Tracking(3DMOT) methods terminate a tracklet when it is not associated with new detections for a few frames. But if an object just goes dark, like being temporarily occluded by other objects or simply getting out of FOV, terminating a tracklet prematurely will result in an identity switch. We reveal that premature tracklet termination is the main cause of identity s… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.13672v1-abstract-full').style.display = 'inline'; document.getElementById('2111.13672v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2111.13672v1-abstract-full" style="display: none;"> Previous online 3D Multi-Object Tracking(3DMOT) methods terminate a tracklet when it is not associated with new detections for a few frames. But if an object just goes dark, like being temporarily occluded by other objects or simply getting out of FOV, terminating a tracklet prematurely will result in an identity switch. We reveal that premature tracklet termination is the main cause of identity switches in modern 3DMOT systems. To address this, we propose Immortal Tracker, a simple tracking system that utilizes trajectory prediction to maintain tracklets for objects gone dark. We employ a simple Kalman filter for trajectory prediction and preserve the tracklet by prediction when the target is not visible. With this method, we can avoid 96% vehicle identity switches resulting from premature tracklet termination. Without any learned parameters, our method achieves a mismatch ratio at the 0.0001 level and competitive MOTA for the vehicle class on the Waymo Open Dataset test set. Our mismatch ratio is tens of times lower than any previously published method. Similar results are reported on nuScenes. We believe the proposed Immortal Tracker can offer a simple yet powerful solution for pushing the limit of 3DMOT. Our code is available at https://github.com/ImmortalTracker/ImmortalTracker. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.13672v1-abstract-full').style.display = 'none'; document.getElementById('2111.13672v1-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> 26 November, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2021. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2111.10586">arXiv:2111.10586</a> <span> [<a href="https://arxiv.org/pdf/2111.10586">pdf</a>, <a href="https://arxiv.org/format/2111.10586">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Signal Processing">eess.SP</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"> Satellite Based Computing Networks with Federated Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chen%2C+H">Hao Chen</a>, <a href="/search/cs?searchtype=author&query=Xiao%2C+M">Ming Xiao</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhibo Pang</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="2111.10586v1-abstract-short" style="display: inline;"> Driven by the ever-increasing penetration and proliferation of data-driven applications, a new generation of wireless communication, the sixth-generation (6G) mobile system enhanced by artificial intelligence (AI), has attracted substantial research interests. Among various candidate technologies of 6G, low earth orbit (LEO) satellites have appealing characteristics of ubiquitous wireless access.… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.10586v1-abstract-full').style.display = 'inline'; document.getElementById('2111.10586v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2111.10586v1-abstract-full" style="display: none;"> Driven by the ever-increasing penetration and proliferation of data-driven applications, a new generation of wireless communication, the sixth-generation (6G) mobile system enhanced by artificial intelligence (AI), has attracted substantial research interests. Among various candidate technologies of 6G, low earth orbit (LEO) satellites have appealing characteristics of ubiquitous wireless access. However, the costs of satellite communication (SatCom) are still high, relative to counterparts of ground mobile networks. To support massively interconnected devices with intelligent adaptive learning and reduce expensive traffic in SatCom, we propose federated learning (FL) in LEO-based satellite communication networks. We first review the state-of-the-art LEO-based SatCom and related machine learning (ML) techniques, and then analyze four possible ways of combining ML with satellite networks. The learning performance of the proposed strategies is evaluated by simulation and results reveal that FL-based computing networks improve the performance of communication overheads and latency. Finally, we discuss future research topics along this research direction. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.10586v1-abstract-full').style.display = 'none'; document.getElementById('2111.10586v1-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, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2021. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2111.09621">arXiv:2111.09621</a> <span> [<a href="https://arxiv.org/pdf/2111.09621">pdf</a>, <a href="https://arxiv.org/format/2111.09621">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"> SimpleTrack: Understanding and Rethinking 3D Multi-object Tracking </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+Z">Zhichao Li</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+N">Naiyan 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="2111.09621v1-abstract-short" style="display: inline;"> 3D multi-object tracking (MOT) has witnessed numerous novel benchmarks and approaches in recent years, especially those under the "tracking-by-detection" paradigm. Despite their progress and usefulness, an in-depth analysis of their strengths and weaknesses is not yet available. In this paper, we summarize current 3D MOT methods into a unified framework by decomposing them into four constituent pa… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.09621v1-abstract-full').style.display = 'inline'; document.getElementById('2111.09621v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2111.09621v1-abstract-full" style="display: none;"> 3D multi-object tracking (MOT) has witnessed numerous novel benchmarks and approaches in recent years, especially those under the "tracking-by-detection" paradigm. Despite their progress and usefulness, an in-depth analysis of their strengths and weaknesses is not yet available. In this paper, we summarize current 3D MOT methods into a unified framework by decomposing them into four constituent parts: pre-processing of detection, association, motion model, and life cycle management. We then ascribe the failure cases of existing algorithms to each component and investigate them in detail. Based on the analyses, we propose corresponding improvements which lead to a strong yet simple baseline: SimpleTrack. Comprehensive experimental results on Waymo Open Dataset and nuScenes demonstrate that our final method could achieve new state-of-the-art results with minor modifications. Furthermore, we take additional steps and rethink whether current benchmarks authentically reflect the ability of algorithms for real-world challenges. We delve into the details of existing benchmarks and find some intriguing facts. Finally, we analyze the distribution and causes of remaining failures in \name\ and propose future directions for 3D MOT. Our code is available at https://github.com/TuSimple/SimpleTrack. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.09621v1-abstract-full').style.display = 'none'; document.getElementById('2111.09621v1-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, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2021. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2111.00695">arXiv:2111.00695</a> <span> [<a href="https://arxiv.org/pdf/2111.00695">pdf</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> </div> </div> <p class="title is-5 mathjax"> Noise Error Pattern Generation Based on Successive Addition-Subtraction for Guessing Decoding </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhan%2C+M">Ming Zhan</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhibo Pang</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+K">Kan Yu</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+J">Jing Xu</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+F">Fang Wu</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="2111.00695v1-abstract-short" style="display: inline;"> Guessing random additive noise decoding (GRAND) algorithm has emerged as an excellent decoding strategy that can meet both the high reliability and low latency constraints. This paper proposes a successive addition-subtraction algorithm to generate noise error permutations. A noise error patterns generation scheme is presented by embedding the "1" and "0" bursts alternately. Then detailed procedur… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.00695v1-abstract-full').style.display = 'inline'; document.getElementById('2111.00695v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2111.00695v1-abstract-full" style="display: none;"> Guessing random additive noise decoding (GRAND) algorithm has emerged as an excellent decoding strategy that can meet both the high reliability and low latency constraints. This paper proposes a successive addition-subtraction algorithm to generate noise error permutations. A noise error patterns generation scheme is presented by embedding the "1" and "0" bursts alternately. Then detailed procedures of the proposed algorithm are presented, and its correctness is also demonstrated through theoretical derivations. The aim of this work is to provide a preliminary paradigm and reference for future research on GRAND algorithm and hardware implementation. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.00695v1-abstract-full').style.display = 'none'; document.getElementById('2111.00695v1-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, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2021. </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">6 pages, 7 figures, submitted to IEEE Communications Letters</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2109.05889">arXiv:2109.05889</a> <span> [<a href="https://arxiv.org/pdf/2109.05889">pdf</a>, <a href="https://arxiv.org/format/2109.05889">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 class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1109/LGRS.2021.3124804">10.1109/LGRS.2021.3124804 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Nonlocal Patch-Based Fully-Connected Tensor Network Decomposition for Remote Sensing Image Inpainting </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zheng%2C+W">Wen-Jie Zheng</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+X">Xi-Le Zhao</a>, <a href="/search/cs?searchtype=author&query=Zheng%2C+Y">Yu-Bang Zheng</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhi-Feng Pang</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="2109.05889v1-abstract-short" style="display: inline;"> Remote sensing image (RSI) inpainting plays an important role in real applications. Recently, fully-connected tensor network (FCTN) decomposition has been shown the remarkable ability to fully characterize the global correlation. Considering the global correlation and the nonlocal self-similarity (NSS) of RSIs, this paper introduces the FCTN decomposition to the whole RSI and its NSS groups, and p… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2109.05889v1-abstract-full').style.display = 'inline'; document.getElementById('2109.05889v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2109.05889v1-abstract-full" style="display: none;"> Remote sensing image (RSI) inpainting plays an important role in real applications. Recently, fully-connected tensor network (FCTN) decomposition has been shown the remarkable ability to fully characterize the global correlation. Considering the global correlation and the nonlocal self-similarity (NSS) of RSIs, this paper introduces the FCTN decomposition to the whole RSI and its NSS groups, and proposes a novel nonlocal patch-based FCTN (NL-FCTN) decomposition for RSI inpainting. Different from other nonlocal patch-based methods, the NL-FCTN decomposition-based method, which increases tensor order by stacking similar small-sized patches to NSS groups, cleverly leverages the remarkable ability of FCTN decomposition to deal with higher-order tensors. Besides, we propose an efficient proximal alternating minimization-based algorithm to solve the proposed NL-FCTN decomposition-based model with a theoretical convergence guarantee. Extensive experiments on RSIs demonstrate that the proposed method achieves the state-of-the-art inpainting performance in all compared methods. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2109.05889v1-abstract-full').style.display = 'none'; document.getElementById('2109.05889v1-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 September, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2021. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> IEEE Geoscience and Remote Sensing Letters, 2021 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2103.11441">arXiv:2103.11441</a> <span> [<a href="https://arxiv.org/pdf/2103.11441">pdf</a>, <a href="https://arxiv.org/format/2103.11441">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"> TextFlint: Unified Multilingual Robustness Evaluation Toolkit for Natural Language Processing </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Gui%2C+T">Tao Gui</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+X">Xiao Wang</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Q">Qi Zhang</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Q">Qin Liu</a>, <a href="/search/cs?searchtype=author&query=Zou%2C+Y">Yicheng Zou</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+X">Xin Zhou</a>, <a href="/search/cs?searchtype=author&query=Zheng%2C+R">Rui Zheng</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+C">Chong Zhang</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+Q">Qinzhuo Wu</a>, <a href="/search/cs?searchtype=author&query=Ye%2C+J">Jiacheng Ye</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zexiong Pang</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Y">Yongxin Zhang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+Z">Zhengyan Li</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+R">Ruotian Ma</a>, <a href="/search/cs?searchtype=author&query=Fei%2C+Z">Zichu Fei</a>, <a href="/search/cs?searchtype=author&query=Cai%2C+R">Ruijian Cai</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+J">Jun Zhao</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+X">Xingwu Hu</a>, <a href="/search/cs?searchtype=author&query=Yan%2C+Z">Zhiheng Yan</a>, <a href="/search/cs?searchtype=author&query=Tan%2C+Y">Yiding Tan</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yuan Hu</a>, <a href="/search/cs?searchtype=author&query=Bian%2C+Q">Qiyuan Bian</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Z">Zhihua Liu</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+B">Bolin Zhu</a>, <a href="/search/cs?searchtype=author&query=Qin%2C+S">Shan Qin</a> , et al. (9 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="2103.11441v3-abstract-short" style="display: inline;"> Various robustness evaluation methodologies from different perspectives have been proposed for different natural language processing (NLP) tasks. These methods have often focused on either universal or task-specific generalization capabilities. In this work, we propose a multilingual robustness evaluation platform for NLP tasks (TextFlint) that incorporates universal text transformation, task-spec… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2103.11441v3-abstract-full').style.display = 'inline'; document.getElementById('2103.11441v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2103.11441v3-abstract-full" style="display: none;"> Various robustness evaluation methodologies from different perspectives have been proposed for different natural language processing (NLP) tasks. These methods have often focused on either universal or task-specific generalization capabilities. In this work, we propose a multilingual robustness evaluation platform for NLP tasks (TextFlint) that incorporates universal text transformation, task-specific transformation, adversarial attack, subpopulation, and their combinations to provide comprehensive robustness analysis. TextFlint enables practitioners to automatically evaluate their models from all aspects or to customize their evaluations as desired with just a few lines of code. To guarantee user acceptability, all the text transformations are linguistically based, and we provide a human evaluation for each one. TextFlint generates complete analytical reports as well as targeted augmented data to address the shortcomings of the model's robustness. To validate TextFlint's utility, we performed large-scale empirical evaluations (over 67,000 evaluations) on state-of-the-art deep learning models, classic supervised methods, and real-world systems. Almost all models showed significant performance degradation, including a decline of more than 50% of BERT's prediction accuracy on tasks such as aspect-level sentiment classification, named entity recognition, and natural language inference. Therefore, we call for the robustness to be included in the model evaluation, so as to promote the healthy development of NLP technology. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2103.11441v3-abstract-full').style.display = 'none'; document.getElementById('2103.11441v3-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 May, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 21 March, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2021. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2103.06028">arXiv:2103.06028</a> <span> [<a href="https://arxiv.org/pdf/2103.06028">pdf</a>, <a href="https://arxiv.org/format/2103.06028">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"> Model-free Vehicle Tracking and State Estimation in Point Cloud Sequences </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+Z">Zhichao Li</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+N">Naiyan 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="2103.06028v2-abstract-short" style="display: inline;"> Estimating the states of surrounding traffic participants stays at the core of autonomous driving. In this paper, we study a novel setting of this problem: model-free single-object tracking (SOT), which takes the object state in the first frame as input, and jointly solves state estimation and tracking in subsequent frames. The main purpose for this new setting is to break the strong limitation of… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2103.06028v2-abstract-full').style.display = 'inline'; document.getElementById('2103.06028v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2103.06028v2-abstract-full" style="display: none;"> Estimating the states of surrounding traffic participants stays at the core of autonomous driving. In this paper, we study a novel setting of this problem: model-free single-object tracking (SOT), which takes the object state in the first frame as input, and jointly solves state estimation and tracking in subsequent frames. The main purpose for this new setting is to break the strong limitation of the popular "detection and tracking" scheme in multi-object tracking. Moreover, we notice that shape completion by overlaying the point clouds, which is a by-product of our proposed task, not only improves the performance of state estimation but also has numerous applications. As no benchmark for this task is available so far, we construct a new dataset LiDAR-SOT and corresponding evaluation protocols based on the Waymo Open dataset. We then propose an optimization-based algorithm called SOTracker involving point cloud registration, vehicle shapes, correspondence, and motion priors. Our quantitative and qualitative results prove the effectiveness of our SOTracker and reveal the challenging cases for SOT in point clouds, including the sparsity of LiDAR data, abrupt motion variation, etc. Finally, we also explore how the proposed task and algorithm may benefit other autonomous driving applications, including simulating LiDAR scans, generating motion data, and annotating optical flow. The code and protocols for our benchmark and algorithm are available at https://github.com/TuSimple/LiDAR_SOT/. A video demonstration is at https://www.youtube.com/watch?v=BpHixKs91i8. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2103.06028v2-abstract-full').style.display = 'none'; document.getElementById('2103.06028v2-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 August, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 10 March, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2021. </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 IROS2021, Camera ready version</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2102.01955">arXiv:2102.01955</a> <span> [<a href="https://arxiv.org/pdf/2102.01955">pdf</a>, <a href="https://arxiv.org/format/2102.01955">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="Neurons and Cognition">q-bio.NC</span> </div> </div> <p class="title is-5 mathjax"> Predictive coding feedback results in perceived illusory contours in a recurrent neural network </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhaoyang Pang</a>, <a href="/search/cs?searchtype=author&query=O%27May%2C+C+B">Callum Biggs O'May</a>, <a href="/search/cs?searchtype=author&query=Choksi%2C+B">Bhavin Choksi</a>, <a href="/search/cs?searchtype=author&query=VanRullen%2C+R">Rufin VanRullen</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="2102.01955v2-abstract-short" style="display: inline;"> Modern feedforward convolutional neural networks (CNNs) can now solve some computer vision tasks at super-human levels. However, these networks only roughly mimic human visual perception. One difference from human vision is that they do not appear to perceive illusory contours (e.g. Kanizsa squares) in the same way humans do. Physiological evidence from visual cortex suggests that the perception o… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2102.01955v2-abstract-full').style.display = 'inline'; document.getElementById('2102.01955v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2102.01955v2-abstract-full" style="display: none;"> Modern feedforward convolutional neural networks (CNNs) can now solve some computer vision tasks at super-human levels. However, these networks only roughly mimic human visual perception. One difference from human vision is that they do not appear to perceive illusory contours (e.g. Kanizsa squares) in the same way humans do. Physiological evidence from visual cortex suggests that the perception of illusory contours could involve feedback connections. Would recurrent feedback neural networks perceive illusory contours like humans? In this work we equip a deep feedforward convolutional network with brain-inspired recurrent dynamics. The network was first pretrained with an unsupervised reconstruction objective on a natural image dataset, to expose it to natural object contour statistics. Then, a classification decision layer was added and the model was finetuned on a form discrimination task: squares vs. randomly oriented inducer shapes (no illusory contour). Finally, the model was tested with the unfamiliar ''illusory contour'' configuration: inducer shapes oriented to form an illusory square. Compared with feedforward baselines, the iterative ''predictive coding'' feedback resulted in more illusory contours being classified as physical squares. The perception of the illusory contour was measurable in the luminance profile of the image reconstructions produced by the model, demonstrating that the model really ''sees'' the illusion. Ablation studies revealed that natural image pretraining and feedback error correction are both critical to the perception of the illusion. Finally we validated our conclusions in a deeper network (VGG): adding the same predictive coding feedback dynamics again leads to the perception of illusory contours. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2102.01955v2-abstract-full').style.display = 'none'; document.getElementById('2102.01955v2-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 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 3 February, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2021. </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">Manuscript under review</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2012.07748">arXiv:2012.07748</a> <span> [<a href="https://arxiv.org/pdf/2012.07748">pdf</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computers and Society">cs.CY</span> </div> </div> <p class="title is-5 mathjax"> Investigation of the Impacts of COVID-19 on the Electricity Consumption of a University Dormitory Using Weather Normalization </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhihong Pang</a>, <a href="/search/cs?searchtype=author&query=Feng%2C+F">Fan Feng</a>, <a href="/search/cs?searchtype=author&query=O%27Neill%2C+Z">Zheng O'Neill</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="2012.07748v1-abstract-short" style="display: inline;"> This study investigated the impacts of the COVID-19 pandemic on the electricity consumption of a university dormitory building in the southern U.S. The historical electricity consumption data of this university dormitory building and weather data of an on-campus weather station, which were collected from January 1st, 2017 to July 31st, 2020, were used for analysis. Four inverse data-driven predict… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2012.07748v1-abstract-full').style.display = 'inline'; document.getElementById('2012.07748v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2012.07748v1-abstract-full" style="display: none;"> This study investigated the impacts of the COVID-19 pandemic on the electricity consumption of a university dormitory building in the southern U.S. The historical electricity consumption data of this university dormitory building and weather data of an on-campus weather station, which were collected from January 1st, 2017 to July 31st, 2020, were used for analysis. Four inverse data-driven prediction models, i.e., Artificial Neural Network, Long Short-Term Memory Recurrent Neural Network, eXtreme Gradient Boosting, and Light Gradient Boosting Machine, were exploited to account for the influence of the weather conditions. The results suggested that the total electricity consumption of the objective building decreased by nearly 41% (about 276,000 kWh (942 MMBtu)) compared with the prediction value during the campus shutdown due to the COVID-19. Besides, the daily load ratio (DLR) varied significantly as well. In general, the DLR decreased gradually from 80% to nearly 40% in the second half of March 2020, maintained on a relatively stable level between 30% to 60% in April, May, and June 2020, and then slowly recovered to 80% of the normal capacity in July 2020. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2012.07748v1-abstract-full').style.display = 'none'; document.getElementById('2012.07748v1-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 December, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2020. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2007.07437">arXiv:2007.07437</a> <span> [<a href="https://arxiv.org/pdf/2007.07437">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"> ContourRend: A Segmentation Method for Improving Contours by Rendering </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chen%2C+J">Junwen Chen</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+Y">Yi Lu</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Y">Yaran Chen</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+D">Dongbin Zhao</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhonghua Pang</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="2007.07437v1-abstract-short" style="display: inline;"> A good object segmentation should contain clear contours and complete regions. However, mask-based segmentation can not handle contour features well on a coarse prediction grid, thus causing problems of blurry edges. While contour-based segmentation provides contours directly, but misses contours' details. In order to obtain fine contours, we propose a segmentation method named ContourRend which a… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2007.07437v1-abstract-full').style.display = 'inline'; document.getElementById('2007.07437v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2007.07437v1-abstract-full" style="display: none;"> A good object segmentation should contain clear contours and complete regions. However, mask-based segmentation can not handle contour features well on a coarse prediction grid, thus causing problems of blurry edges. While contour-based segmentation provides contours directly, but misses contours' details. In order to obtain fine contours, we propose a segmentation method named ContourRend which adopts a contour renderer to refine segmentation contours. And we implement our method on a segmentation model based on graph convolutional network (GCN). For the single object segmentation task on cityscapes dataset, the GCN-based segmentation con-tour is used to generate a contour of a single object, then our contour renderer focuses on the pixels around the contour and predicts the category at high resolution. By rendering the contour result, our method reaches 72.41% mean intersection over union (IoU) and surpasses baseline Polygon-GCN by 1.22%. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2007.07437v1-abstract-full').style.display = 'none'; document.getElementById('2007.07437v1-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 July, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2020. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2003.11941">arXiv:2003.11941</a> <span> [<a href="https://arxiv.org/pdf/2003.11941">pdf</a>, <a href="https://arxiv.org/format/2003.11941">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> <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"> AliExpress Learning-To-Rank: Maximizing Online Model Performance without Going Online </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Huzhang%2C+G">Guangda Huzhang</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhen-Jia Pang</a>, <a href="/search/cs?searchtype=author&query=Gao%2C+Y">Yongqing Gao</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Y">Yawen Liu</a>, <a href="/search/cs?searchtype=author&query=Shen%2C+W">Weijie Shen</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+W">Wen-Ji Zhou</a>, <a href="/search/cs?searchtype=author&query=Da%2C+Q">Qing Da</a>, <a href="/search/cs?searchtype=author&query=Zeng%2C+A">An-Xiang Zeng</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+H">Han Yu</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+Y">Yang Yu</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+Z">Zhi-Hua Zhou</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="2003.11941v5-abstract-short" style="display: inline;"> Learning-to-rank (LTR) has become a key technology in E-commerce applications. Most existing LTR approaches follow a supervised learning paradigm from offline labeled data collected from the online system. However, it has been noticed that previous LTR models can have a good validation performance over offline validation data but have a poor online performance, and vice versa, which implies a poss… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2003.11941v5-abstract-full').style.display = 'inline'; document.getElementById('2003.11941v5-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2003.11941v5-abstract-full" style="display: none;"> Learning-to-rank (LTR) has become a key technology in E-commerce applications. Most existing LTR approaches follow a supervised learning paradigm from offline labeled data collected from the online system. However, it has been noticed that previous LTR models can have a good validation performance over offline validation data but have a poor online performance, and vice versa, which implies a possible large inconsistency between the offline and online evaluation. We investigate and confirm in this paper that such inconsistency exists and can have a significant impact on AliExpress Search. Reasons for the inconsistency include the ignorance of item context during the learning, and the offline data set is insufficient for learning the context. Therefore, this paper proposes an evaluator-generator framework for LTR with item context. The framework consists of an evaluator that generalizes to evaluate recommendations involving the context, and a generator that maximizes the evaluator score by reinforcement learning, and a discriminator that ensures the generalization of the evaluator. Extensive experiments in simulation environments and AliExpress Search online system show that, firstly, the classic data-based metrics on the offline dataset can show significant inconsistency with online performance, and can even be misleading. Secondly, the proposed evaluator score is significantly more consistent with the online performance than common ranking metrics. Finally, as the consequence, our method achieves a significant improvement (\textgreater$2\%$) in terms of Conversion Rate (CR) over the industrial-level fine-tuned model in online A/B tests. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2003.11941v5-abstract-full').style.display = 'none'; document.getElementById('2003.11941v5-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 December, 2020; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 25 March, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2020. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1912.07186">arXiv:1912.07186</a> <span> [<a href="https://arxiv.org/pdf/1912.07186">pdf</a>, <a href="https://arxiv.org/format/1912.07186">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Signal Processing">eess.SP</span> </div> </div> <p class="title is-5 mathjax"> Minimizing Age of Information for Real-Time Monitoring in Resource-Constrained Industrial IoT Networks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wang%2C+Q">Qian Wang</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+H">He Chen</a>, <a href="/search/cs?searchtype=author&query=Li%2C+Y">Yonghui Li</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhibo Pang</a>, <a href="/search/cs?searchtype=author&query=Vucetic%2C+B">Branka Vucetic</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="1912.07186v1-abstract-short" style="display: inline;"> This paper considers an Industrial Internet of Thing (IIoT) system with a source monitoring a dynamic process with randomly generated status updates. The status updates are sent to an designated destination in a real-time manner over an unreliable link. The source is subject to a practical constraint of limited average transmission power. Thus, the system should carefully schedule when to transmit… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1912.07186v1-abstract-full').style.display = 'inline'; document.getElementById('1912.07186v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1912.07186v1-abstract-full" style="display: none;"> This paper considers an Industrial Internet of Thing (IIoT) system with a source monitoring a dynamic process with randomly generated status updates. The status updates are sent to an designated destination in a real-time manner over an unreliable link. The source is subject to a practical constraint of limited average transmission power. Thus, the system should carefully schedule when to transmit a fresh status update or retransmit the stale one. To characterize the performance of timely status update, we adopt a recent concept, Age of Information (AoI), as the performance metric. We aim to minimize the long-term average AoI under the limited average transmission power at the source, by formulating a constrained Markov Decision Process (CMDP) problem. To address the formulated CMDP, we recast it into an unconstrained Markov Decision Process (MDP) through Lagrangian relaxation. We prove the existence of optimal stationary policy of the original CMDP, which is a randomized mixture of two deterministic stationary policies of the unconstrained MDP. We also explore the characteristics of the problem to reduce the action space of each state to significantly reduce the computation complexity. We further prove the threshold structure of the optimal deterministic policy for the unconstrained MDP. Simulation results show the proposed optimal policy achieves lower average AoI compared with random policy, especially when the system suffers from stricter resource constraint. Besides, the influence of status generation probability and transmission failure rate on optimal policy and the resultant average AoI as well as the impact of average transmission power on the minimal average AoI are unveiled. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1912.07186v1-abstract-full').style.display = 'none'; document.getElementById('1912.07186v1-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 December, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2019. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1911.12911">arXiv:1911.12911</a> <span> [<a href="https://arxiv.org/pdf/1911.12911">pdf</a>, <a href="https://arxiv.org/format/1911.12911">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"> Unlocking the Full Potential of Small Data with Diverse Supervision </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Ziqi Pang</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+Z">Zhiyuan Hu</a>, <a href="/search/cs?searchtype=author&query=Tokmakov%2C+P">Pavel Tokmakov</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yu-Xiong Wang</a>, <a href="/search/cs?searchtype=author&query=Hebert%2C+M">Martial Hebert</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="1911.12911v3-abstract-short" style="display: inline;"> Virtually all of deep learning literature relies on the assumption of large amounts of available training data. Indeed, even the majority of few-shot learning methods rely on a large set of "base classes" for pretraining. This assumption, however, does not always hold. For some tasks, annotating a large number of classes can be infeasible, and even collecting the images themselves can be a challen… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1911.12911v3-abstract-full').style.display = 'inline'; document.getElementById('1911.12911v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1911.12911v3-abstract-full" style="display: none;"> Virtually all of deep learning literature relies on the assumption of large amounts of available training data. Indeed, even the majority of few-shot learning methods rely on a large set of "base classes" for pretraining. This assumption, however, does not always hold. For some tasks, annotating a large number of classes can be infeasible, and even collecting the images themselves can be a challenge in some scenarios. In this paper, we study this problem and call it "Small Data" setting, in contrast to "Big Data". To unlock the full potential of small data, we propose to augment the models with annotations for other related tasks, thus increasing their generalization abilities. In particular, we use the richly annotated scene parsing dataset ADE20K to construct our realistic Long-tail Recognition with Diverse Supervision (LRDS) benchmark by splitting the object categories into head and tail based on their distribution. Following the standard few-shot learning protocol, we use the head classes for representation learning and the tail classes for evaluation. Moreover, we further subsample the head categories and images to generate two novel settings which we call "Scarce-Class" and "Scarce-Image", respectively corresponding to the shortage of samples for rare classes and training images. Finally, we analyze the effect of applying various additional supervision sources under the proposed settings. Our experiments demonstrate that densely labeling a small set of images can indeed largely remedy the small data constraints. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1911.12911v3-abstract-full').style.display = 'none'; document.getElementById('1911.12911v3-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> 26 April, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 28 November, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2019. </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">Learning from Limited and Imperfect Data (L2ID) Workshop @ CVPR 2021</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1903.00715">arXiv:1903.00715</a> <span> [<a href="https://arxiv.org/pdf/1903.00715">pdf</a>, <a href="https://arxiv.org/format/1903.00715">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> <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"> Efficient Reinforcement Learning for StarCraft by Abstract Forward Models and Transfer Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liu%2C+R">Ruo-Ze Liu</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+H">Haifeng Guo</a>, <a href="/search/cs?searchtype=author&query=Ji%2C+X">Xiaozhong Ji</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+Y">Yang Yu</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhen-Jia Pang</a>, <a href="/search/cs?searchtype=author&query=Xiao%2C+Z">Zitai Xiao</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+Y">Yuzhou Wu</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+T">Tong 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="1903.00715v4-abstract-short" style="display: inline;"> Injecting human knowledge is an effective way to accelerate reinforcement learning (RL). However, these methods are underexplored. This paper presents our discovery that an abstract forward model (thought-game (TG)) combined with transfer learning (TL) is an effective way. We take StarCraft II as our study environment. With the help of a designed TG, the agent can learn a 99% win-rate on a 64x64 m… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1903.00715v4-abstract-full').style.display = 'inline'; document.getElementById('1903.00715v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1903.00715v4-abstract-full" style="display: none;"> Injecting human knowledge is an effective way to accelerate reinforcement learning (RL). However, these methods are underexplored. This paper presents our discovery that an abstract forward model (thought-game (TG)) combined with transfer learning (TL) is an effective way. We take StarCraft II as our study environment. With the help of a designed TG, the agent can learn a 99% win-rate on a 64x64 map against the Level-7 built-in AI, using only 1.08 hours in a single commercial machine. We also show that the TG method is not as restrictive as it was thought to be. It can work with roughly designed TGs, and can also be useful when the environment changes. Comparing with previous model-based RL, we show TG is more effective. We also present a TG hypothesis that gives the influence of different fidelity levels of TG. For real games that have unequal state and action spaces, we proposed a novel XfrNet of which usefulness is validated while achieving a 90% win-rate against the cheating Level-10 AI. We argue that the TG method might shed light on further studies of efficient RL with human knowledge. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1903.00715v4-abstract-full').style.display = 'none'; document.getElementById('1903.00715v4-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> 2 November, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 2 March, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2019. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1811.06166">arXiv:1811.06166</a> <span> [<a href="https://arxiv.org/pdf/1811.06166">pdf</a>, <a href="https://arxiv.org/format/1811.06166">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Multimedia">cs.MM</span> </div> </div> <p class="title is-5 mathjax"> Tiyuntsong: A Self-Play Reinforcement Learning Approach for ABR Video Streaming </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Huang%2C+T">Tianchi Huang</a>, <a href="/search/cs?searchtype=author&query=Yao%2C+X">Xin Yao</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+C">Chenglei Wu</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+R">Rui-Xiao Zhang</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhangyuan Pang</a>, <a href="/search/cs?searchtype=author&query=Sun%2C+L">Lifeng 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="1811.06166v3-abstract-short" style="display: inline;"> Existing reinforcement learning~(RL)-based adaptive bitrate~(ABR) approaches outperform the previous fixed control rules based methods by improving the Quality of Experience~(QoE) score, as the QoE metric can hardly provide clear guidance for optimization, finally resulting in the unexpected strategies. In this paper, we propose \emph{Tiyuntsong}, a self-play reinforcement learning approach with g… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1811.06166v3-abstract-full').style.display = 'inline'; document.getElementById('1811.06166v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1811.06166v3-abstract-full" style="display: none;"> Existing reinforcement learning~(RL)-based adaptive bitrate~(ABR) approaches outperform the previous fixed control rules based methods by improving the Quality of Experience~(QoE) score, as the QoE metric can hardly provide clear guidance for optimization, finally resulting in the unexpected strategies. In this paper, we propose \emph{Tiyuntsong}, a self-play reinforcement learning approach with generative adversarial network~(GAN)-based method for ABR video streaming. Tiyuntsong learns strategies automatically by training two agents who are competing against each other. Note that the competition results are determined by a set of rules rather than a numerical QoE score that allows clearer optimization objectives. Meanwhile, we propose GAN Enhancement Module to extract hidden features from the past status for preserving the information without the limitations of sequence lengths. Using testbed experiments, we show that the utilization of GAN significantly improves the Tiyuntsong's performance. By comparing the performance of ABRs, we observe that Tiyuntsong also betters existing ABR algorithms in the underlying metrics. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1811.06166v3-abstract-full').style.display = 'none'; document.getElementById('1811.06166v3-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> 2 May, 2019; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 November, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2018. </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">Published in ICME 2019</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1809.09095">arXiv:1809.09095</a> <span> [<a href="https://arxiv.org/pdf/1809.09095">pdf</a>, <a href="https://arxiv.org/format/1809.09095">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> <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"> On Reinforcement Learning for Full-length Game of StarCraft </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhen-Jia Pang</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+R">Ruo-Ze Liu</a>, <a href="/search/cs?searchtype=author&query=Meng%2C+Z">Zhou-Yu Meng</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Y">Yi Zhang</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+Y">Yang Yu</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+T">Tong 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="1809.09095v2-abstract-short" style="display: inline;"> StarCraft II poses a grand challenge for reinforcement learning. The main difficulties of it include huge state and action space and a long-time horizon. In this paper, we investigate a hierarchical reinforcement learning approach for StarCraft II. The hierarchy involves two levels of abstraction. One is the macro-action automatically extracted from expert's trajectories, which reduces the action… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1809.09095v2-abstract-full').style.display = 'inline'; document.getElementById('1809.09095v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1809.09095v2-abstract-full" style="display: none;"> StarCraft II poses a grand challenge for reinforcement learning. The main difficulties of it include huge state and action space and a long-time horizon. In this paper, we investigate a hierarchical reinforcement learning approach for StarCraft II. The hierarchy involves two levels of abstraction. One is the macro-action automatically extracted from expert's trajectories, which reduces the action space in an order of magnitude yet remains effective. The other is a two-layer hierarchical architecture which is modular and easy to scale, enabling a curriculum transferring from simpler tasks to more complex tasks. The reinforcement training algorithm for this architecture is also investigated. On a 64x64 map and using restrictive units, we achieve a winning rate of more than 99\% against the difficulty level-1 built-in AI. Through the curriculum transfer learning algorithm and a mixture of combat model, we can achieve over 93\% winning rate of Protoss against the most difficult non-cheating built-in AI (level-7) of Terran, training within two days using a single machine with only 48 CPU cores and 8 K40 GPUs. It also shows strong generalization performance, when tested against never seen opponents including cheating levels built-in AI and all levels of Zerg and Protoss built-in AI. We hope this study could shed some light on the future research of large-scale reinforcement learning. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1809.09095v2-abstract-full').style.display = 'none'; document.getElementById('1809.09095v2-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 February, 2019; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 23 September, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2018. </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">Appeared in AAAI 2019</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1808.02079">arXiv:1808.02079</a> <span> [<a href="https://arxiv.org/pdf/1808.02079">pdf</a>, <a href="https://arxiv.org/format/1808.02079">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Signal Processing">eess.SP</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Networking and Internet Architecture">cs.NI</span> </div> </div> <p class="title is-5 mathjax"> Low-latency Networking: Where Latency Lurks and How to Tame It </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jiang%2C+X">Xiaolin Jiang</a>, <a href="/search/cs?searchtype=author&query=Ghadikolaei%2C+H+S">Hossein S. Ghadikolaei</a>, <a href="/search/cs?searchtype=author&query=Fodor%2C+G">Gabor Fodor</a>, <a href="/search/cs?searchtype=author&query=Modiano%2C+E">Eytan Modiano</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhibo Pang</a>, <a href="/search/cs?searchtype=author&query=Zorzi%2C+M">Michele Zorzi</a>, <a href="/search/cs?searchtype=author&query=Fischione%2C+C">Carlo Fischione</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="1808.02079v1-abstract-short" style="display: inline;"> While the current generation of mobile and fixed communication networks has been standardized for mobile broadband services, the next generation is driven by the vision of the Internet of Things and mission critical communication services requiring latency in the order of milliseconds or sub-milliseconds. However, these new stringent requirements have a large technical impact on the design of all… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1808.02079v1-abstract-full').style.display = 'inline'; document.getElementById('1808.02079v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1808.02079v1-abstract-full" style="display: none;"> While the current generation of mobile and fixed communication networks has been standardized for mobile broadband services, the next generation is driven by the vision of the Internet of Things and mission critical communication services requiring latency in the order of milliseconds or sub-milliseconds. However, these new stringent requirements have a large technical impact on the design of all layers of the communication protocol stack. The cross layer interactions are complex due to the multiple design principles and technologies that contribute to the layers' design and fundamental performance limitations. We will be able to develop low-latency networks only if we address the problem of these complex interactions from the new point of view of sub-milliseconds latency. In this article, we propose a holistic analysis and classification of the main design principles and enabling technologies that will make it possible to deploy low-latency wireless communication networks. We argue that these design principles and enabling technologies must be carefully orchestrated to meet the stringent requirements and to manage the inherent trade-offs between low latency and traditional performance metrics. We also review currently ongoing standardization activities in prominent standards associations, and discuss open problems for future research. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1808.02079v1-abstract-full').style.display = 'none'; document.getElementById('1808.02079v1-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 August, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2018. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1702.06700">arXiv:1702.06700</a> <span> [<a href="https://arxiv.org/pdf/1702.06700">pdf</a>, <a href="https://arxiv.org/format/1702.06700">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> <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="Neural and Evolutionary Computing">cs.NE</span> </div> </div> <p class="title is-5 mathjax"> Task-driven Visual Saliency and Attention-based Visual Question Answering </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lin%2C+Y">Yuetan Lin</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhangyang Pang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+D">Donghui Wang</a>, <a href="/search/cs?searchtype=author&query=Zhuang%2C+Y">Yueting Zhuang</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="1702.06700v1-abstract-short" style="display: inline;"> Visual question answering (VQA) has witnessed great progress since May, 2015 as a classic problem unifying visual and textual data into a system. Many enlightening VQA works explore deep into the image and question encodings and fusing methods, of which attention is the most effective and infusive mechanism. Current attention based methods focus on adequate fusion of visual and textual features, b… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1702.06700v1-abstract-full').style.display = 'inline'; document.getElementById('1702.06700v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1702.06700v1-abstract-full" style="display: none;"> Visual question answering (VQA) has witnessed great progress since May, 2015 as a classic problem unifying visual and textual data into a system. Many enlightening VQA works explore deep into the image and question encodings and fusing methods, of which attention is the most effective and infusive mechanism. Current attention based methods focus on adequate fusion of visual and textual features, but lack the attention to where people focus to ask questions about the image. Traditional attention based methods attach a single value to the feature at each spatial location, which losses many useful information. To remedy these problems, we propose a general method to perform saliency-like pre-selection on overlapped region features by the interrelation of bidirectional LSTM (BiLSTM), and use a novel element-wise multiplication based attention method to capture more competent correlation information between visual and textual features. We conduct experiments on the large-scale COCO-VQA dataset and analyze the effectiveness of our model demonstrated by strong empirical results. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1702.06700v1-abstract-full').style.display = 'none'; document.getElementById('1702.06700v1-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> 22 February, 2017; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2017. </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">8 pages, 3 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/1605.09116">arXiv:1605.09116</a> <span> [<a href="https://arxiv.org/pdf/1605.09116">pdf</a>, <a href="https://arxiv.org/ps/1605.09116">ps</a>, <a href="https://arxiv.org/format/1605.09116">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="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Image segmentation based on the hybrid total variation model and the K-means clustering strategy </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shi%2C+B">Baoli Shi</a>, <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhi-Feng Pang</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+J">Jing 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="1605.09116v1-abstract-short" style="display: inline;"> The performance of image segmentation highly relies on the original inputting image. When the image is contaminated by some noises or blurs, we can not obtain the efficient segmentation result by using direct segmentation methods. In order to efficiently segment the contaminated image, this paper proposes a two step method based on the hybrid total variation model with a box constraint and the K-m… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1605.09116v1-abstract-full').style.display = 'inline'; document.getElementById('1605.09116v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1605.09116v1-abstract-full" style="display: none;"> The performance of image segmentation highly relies on the original inputting image. When the image is contaminated by some noises or blurs, we can not obtain the efficient segmentation result by using direct segmentation methods. In order to efficiently segment the contaminated image, this paper proposes a two step method based on the hybrid total variation model with a box constraint and the K-means clustering method. In the first step, the hybrid model is based on the weighted convex combination between the total variation functional and the high-order total variation as the regularization term to obtain the original clustering data. In order to deal with non-smooth regularization term, we solve this model by employing the alternating split Bregman method. Then, in the second step, the segmentation can be obtained by thresholding this clustering data into different phases, where the thresholds can be given by using the K-means clustering method. Numerical comparisons show that our proposed model can provide more efficient segmentation results dealing with the noise image and blurring image. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1605.09116v1-abstract-full').style.display = 'none'; document.getElementById('1605.09116v1-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> 30 May, 2016; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2016. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1509.07211">arXiv:1509.07211</a> <span> [<a href="https://arxiv.org/pdf/1509.07211">pdf</a>, <a href="https://arxiv.org/format/1509.07211">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Sound">cs.SD</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> </div> </div> <p class="title is-5 mathjax"> Noise-Robust ASR for the third 'CHiME' Challenge Exploiting Time-Frequency Masking based Multi-Channel Speech Enhancement and Recurrent Neural Network </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zaihu Pang</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+F">Fengyun Zhu</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="1509.07211v1-abstract-short" style="display: inline;"> In this paper, the Lingban entry to the third 'CHiME' speech separation and recognition challenge is presented. A time-frequency masking based speech enhancement front-end is proposed to suppress the environmental noise utilizing multi-channel coherence and spatial cues. The state-of-the-art speech recognition techniques, namely recurrent neural network based acoustic and language modeling, state… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1509.07211v1-abstract-full').style.display = 'inline'; document.getElementById('1509.07211v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1509.07211v1-abstract-full" style="display: none;"> In this paper, the Lingban entry to the third 'CHiME' speech separation and recognition challenge is presented. A time-frequency masking based speech enhancement front-end is proposed to suppress the environmental noise utilizing multi-channel coherence and spatial cues. The state-of-the-art speech recognition techniques, namely recurrent neural network based acoustic and language modeling, state space minimum Bayes risk based discriminative acoustic modeling, and i-vector based acoustic condition modeling, are carefully integrated into the speech recognition back-end. To further improve the system performance by fully exploiting the advantages of different technologies, the final recognition results are obtained by lattice combination and rescoring. Evaluations carried out on the official dataset prove the effectiveness of the proposed systems. Comparing with the best baseline result, the proposed system obtains consistent improvements with over 57% relative word error rate reduction on the real-data test set. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1509.07211v1-abstract-full').style.display = 'none'; document.getElementById('1509.07211v1-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> 23 September, 2015; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2015. </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">The 3rd 'CHiME' Speech Separation and Recognition Challenge, 5 pages, 1 figure</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1110.1804">arXiv:1110.1804</a> <span>  </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="Information Theory">cs.IT</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"> The proximal point method for a hybrid model in image restoration </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pang%2C+Z">Zhi-Feng Pang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+L">Li-Lian Wang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+Y">Yu-Fei 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="1110.1804v2-abstract-short" style="display: inline;"> Models including two $L^1$ -norm terms have been widely used in image restoration. In this paper we first propose the alternating direction method of multipliers (ADMM) to solve this class of models. Based on ADMM, we then propose the proximal point method (PPM), which is more efficient than ADMM. Following the operator theory, we also give the convergence analysis of the proposed methods. Further… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1110.1804v2-abstract-full').style.display = 'inline'; document.getElementById('1110.1804v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1110.1804v2-abstract-full" style="display: none;"> Models including two $L^1$ -norm terms have been widely used in image restoration. In this paper we first propose the alternating direction method of multipliers (ADMM) to solve this class of models. Based on ADMM, we then propose the proximal point method (PPM), which is more efficient than ADMM. Following the operator theory, we also give the convergence analysis of the proposed methods. Furthermore, we use the proposed methods to solve a class of hybrid models combining the ROF model with the LLT model. Some numerical results demonstrate the viability and efficiency of the proposed methods. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1110.1804v2-abstract-full').style.display = 'none'; document.getElementById('1110.1804v2-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> 25 August, 2012; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 9 October, 2011; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2011. </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">Since we find that there are some unsuitale errors, I withdraw this paper from this website!</span> </p> </li> </ol> <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