Search | arXiv e-print repository

<!DOCTYPE html> <html lang="en"> <head> <meta charset="utf-8"/> <meta name="viewport" content="width=device-width, initial-scale=1"/>  <link rel="apple-touch-icon" sizes="180x180" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/apple-touch-icon.png"> <link rel="icon" type="image/png" sizes="32x32" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon-32x32.png"> <link rel="icon" type="image/png" sizes="16x16" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon-16x16.png"> <link rel="manifest" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/site.webmanifest"> <link rel="mask-icon" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/safari-pinned-tab.svg" color="#b31b1b"> <link rel="shortcut icon" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon.ico"> <meta name="msapplication-TileColor" content="#b31b1b"> <meta name="msapplication-config" content="images/icons/browserconfig.xml"> <meta name="theme-color" content="#b31b1b">  <title>Search | arXiv e-print repository</title> <script defer src="https://static.arxiv.org/static/base/1.0.0a5/fontawesome-free-5.11.2-web/js/all.js"></script> <link rel="stylesheet" href="https://static.arxiv.org/static/base/1.0.0a5/css/arxivstyle.css" /> <script type="text/x-mathjax-config"> MathJax.Hub.Config({ messageStyle: "none", extensions: ["tex2jax.js"], jax: ["input/TeX", "output/HTML-CSS"], tex2jax: { inlineMath: [ ['$','$'], ["\$","\$"] ], displayMath: [ ['$$','$$'], ["\\[","\\]"] ], processEscapes: true, ignoreClass: '.*', processClass: 'mathjax.*' }, TeX: { extensions: ["AMSmath.js", "AMSsymbols.js", "noErrors.js"], noErrors: { inlineDelimiters: ["$","$"], multiLine: false, style: { "font-size": "normal", "border": "" } } }, "HTML-CSS": { availableFonts: ["TeX"] } }); </script> <script src='//static.arxiv.org/MathJax-2.7.3/MathJax.js'></script> <script src="https://static.arxiv.org/static/base/1.0.0a5/js/notification.js"></script> <link rel="stylesheet" href="https://static.arxiv.org/static/search/0.5.6/css/bulma-tooltip.min.css" /> <link rel="stylesheet" href="https://static.arxiv.org/static/search/0.5.6/css/search.css" /> <script src="https://code.jquery.com/jquery-3.2.1.slim.min.js" integrity="sha256-k2WSCIexGzOj3Euiig+TlR8gA0EmPjuc79OEeY5L45g=" crossorigin="anonymous"></script> <script src="https://static.arxiv.org/static/search/0.5.6/js/fieldset.js"></script> <style> radio#cf-customfield_11400 { display: none; } </style> </head> <body> <header><a href="#main-container" class="is-sr-only">Skip to main content</a>  <div class="attribution level is-marginless" role="banner"> <div class="level-left"> <a class="level-item" href="https://cornell.edu/"><img src="https://static.arxiv.org/static/base/1.0.0a5/images/cornell-reduced-white-SMALL.svg" alt="Cornell University" width="200" aria-label="logo" /></a> </div> <div class="level-right is-marginless"><p class="sponsors level-item is-marginless"><span id="support-ack-url">We gratefully acknowledge support from<br /> the Simons Foundation, <a href="https://info.arxiv.org/about/ourmembers.html">member institutions</a>, and all contributors. <a href="https://info.arxiv.org/about/donate.html">Donate</a></span></p></div> </div>  <div class="identity level is-marginless"> <div class="level-left"> <div class="level-item"> <a class="arxiv" href="https://arxiv.org/" aria-label="arxiv-logo"> <img src="https://static.arxiv.org/static/base/1.0.0a5/images/arxiv-logo-one-color-white.svg" aria-label="logo" alt="arxiv logo" width="85" style="width:85px;"/> </a> </div> </div> <div class="search-block level-right"> <form class="level-item mini-search" method="GET" action="https://arxiv.org/search"> <div class="field has-addons"> <div class="control"> <input class="input is-small" type="text" name="query" placeholder="Search..." aria-label="Search term or terms" /> <p class="help"><a href="https://info.arxiv.org/help">Help</a> | <a href="https://arxiv.org/search/advanced">Advanced Search</a></p> </div> <div class="control"> <div class="select is-small"> <select name="searchtype" aria-label="Field to search"> <option value="all" selected="selected">All fields</option> <option value="title">Title</option> <option value="author">Author</option> <option value="abstract">Abstract</option> <option value="comments">Comments</option> <option value="journal_ref">Journal reference</option> <option value="acm_class">ACM classification</option> <option value="msc_class">MSC classification</option> <option value="report_num">Report number</option> <option value="paper_id">arXiv identifier</option> <option value="doi">DOI</option> <option value="orcid">ORCID</option> <option value="author_id">arXiv author ID</option> <option value="help">Help pages</option> <option value="full_text">Full text</option> </select> </div> </div> <input type="hidden" name="source" value="header"> <button class="button is-small is-cul-darker">Search</button> </div> </form> </div> </div>  <div class="container"> <div class="user-tools is-size-7 has-text-right has-text-weight-bold" role="navigation" aria-label="User menu"> <a href="https://arxiv.org/login">Login</a> </div> </div> </header> <main class="container" id="main-container"> <div class="level is-marginless"> <div class="level-left"> <h1 class="title is-clearfix"> Showing 1–50 of 57 results for author: <span class="mathjax">Ostendorf, M</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=Ostendorf%2C+M">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="Ostendorf, M"> </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=Ostendorf%2C+M&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="Ostendorf, M"> <ul id="abstracts"><li><input checked id="abstracts-0" name="abstracts" type="radio" value="show"> <label for="abstracts-0">Show abstracts</label></li><li><input id="abstracts-1" name="abstracts" type="radio" value="hide"> <label for="abstracts-1">Hide abstracts</label></li></ul> </div> <div class="box field is-grouped is-grouped-multiline level-item"> <div class="control"> <span class="select is-small"> <select id="size" name="size"><option value="25">25</option><option selected value="50">50</option><option value="100">100</option><option value="200">200</option></select> </span> <label for="size">results per page</label>. </div> <div class="control"> <label for="order">Sort results by</label> <span class="select is-small"> <select id="order" name="order"><option selected value="-announced_date_first">Announcement date (newest first)</option><option value="announced_date_first">Announcement date (oldest first)</option><option value="-submitted_date">Submission date (newest first)</option><option value="submitted_date">Submission date (oldest first)</option><option value="">Relevance</option></select> </span> </div> <div class="control"> <button class="button is-small is-link">Go</button> </div> </div> </form> </div> </div> <nav class="pagination is-small is-centered breathe-horizontal" role="navigation" aria-label="pagination"> <a href="" class="pagination-previous is-invisible">Previous </a> <a href="/search/?searchtype=author&query=Ostendorf%2C+M&start=50" class="pagination-next" >Next </a> <ul class="pagination-list"> <li> <a href="/search/?searchtype=author&query=Ostendorf%2C+M&start=0" class="pagination-link is-current" aria-label="Goto page 1">1 </a> </li> <li> <a href="/search/?searchtype=author&query=Ostendorf%2C+M&start=50" class="pagination-link " aria-label="Page 2" aria-current="page">2 </a> </li> </ul> </nav> <ol class="breathe-horizontal" start="1"> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2410.18209">arXiv:2410.18209</a> <span> [<a href="https://arxiv.org/pdf/2410.18209">pdf</a>, <a href="https://arxiv.org/format/2410.18209">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"> CorrectionLM: Self-Corrections with SLM for Dialogue State Tracking </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lee%2C+C">Chia-Hsuan Lee</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+H">Hao Cheng</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2410.18209v1-abstract-short" style="display: inline;"> Large language models (LLMs) have demonstrated self-improvement capabilities via feedback and refinement, but current small language models (SLMs) have had limited success in this area. Existing correction approaches often rely on distilling knowledge from LLMs, which imposes significant computation demands. In this work, we introduce CORRECTIONLM, a novel correction framework that enables SLMs to… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.18209v1-abstract-full').style.display = 'inline'; document.getElementById('2410.18209v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.18209v1-abstract-full" style="display: none;"> Large language models (LLMs) have demonstrated self-improvement capabilities via feedback and refinement, but current small language models (SLMs) have had limited success in this area. Existing correction approaches often rely on distilling knowledge from LLMs, which imposes significant computation demands. In this work, we introduce CORRECTIONLM, a novel correction framework that enables SLMs to self-correct using in-context exemplars without LLM involvement. Applied to two dialogue state tracking (DST) tasks in low-resource settings, CORRECTIONLM achieves results similar to a state-of-the-art LLM at a small fraction of the computation costs. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.18209v1-abstract-full').style.display = 'none'; document.getElementById('2410.18209v1-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 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2409.04927">arXiv:2409.04927</a> <span> [<a href="https://arxiv.org/pdf/2409.04927">pdf</a>, <a href="https://arxiv.org/format/2409.04927">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="Audio and Speech Processing">eess.AS</span> </div> </div> <p class="title is-5 mathjax"> Just ASR + LLM? A Study on Speech Large Language Models' Ability to Identify and Understand Speaker in Spoken Dialogue </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wu%2C+J">Junkai Wu</a>, <a href="/search/cs?searchtype=author&query=Fan%2C+X">Xulin Fan</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+B">Bo-Ru Lu</a>, <a href="/search/cs?searchtype=author&query=Jiang%2C+X">Xilin Jiang</a>, <a href="/search/cs?searchtype=author&query=Mesgarani%2C+N">Nima Mesgarani</a>, <a href="/search/cs?searchtype=author&query=Hasegawa-Johnson%2C+M">Mark Hasegawa-Johnson</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2409.04927v3-abstract-short" style="display: inline;"> In recent years, we have observed a rapid advancement in speech language models (SpeechLLMs), catching up with humans' listening and reasoning abilities. SpeechLLMs have demonstrated impressive spoken dialog question-answering (SQA) performance in benchmarks like Gaokao, the English listening test of the college entrance exam in China, which seemingly requires understanding both the spoken content… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.04927v3-abstract-full').style.display = 'inline'; document.getElementById('2409.04927v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2409.04927v3-abstract-full" style="display: none;"> In recent years, we have observed a rapid advancement in speech language models (SpeechLLMs), catching up with humans' listening and reasoning abilities. SpeechLLMs have demonstrated impressive spoken dialog question-answering (SQA) performance in benchmarks like Gaokao, the English listening test of the college entrance exam in China, which seemingly requires understanding both the spoken content and voice characteristics of speakers in a conversation. However, after carefully examining Gaokao's questions, we find the correct answers to many questions can be inferred from the conversation transcript alone, i.e.\ without speaker segmentation and identification. Our evaluation of state-of-the-art models Qwen-Audio and WavLLM on both Gaokao and our proposed "What Do You Like?" dataset shows a significantly higher accuracy in these context-based questions than in identity-critical questions, which can only be answered reliably with correct speaker identification. The results and analysis suggest that when solving SQA, the current SpeechLLMs exhibit limited speaker awareness from the audio and behave similarly to an LLM reasoning from the conversation transcription without sound. We propose that tasks focused on identity-critical questions could offer a more accurate evaluation framework of SpeechLLMs in SQA. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.04927v3-abstract-full').style.display = 'none'; document.getElementById('2409.04927v3-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, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 7 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">Accepted to IEEE SLT 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/2406.09403">arXiv:2406.09403</a> <span> [<a href="https://arxiv.org/pdf/2406.09403">pdf</a>, <a href="https://arxiv.org/format/2406.09403">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="Computation and Language">cs.CL</span> </div> </div> <p class="title is-5 mathjax"> Visual Sketchpad: Sketching as a Visual Chain of Thought for Multimodal Language Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yushi Hu</a>, <a href="/search/cs?searchtype=author&query=Shi%2C+W">Weijia Shi</a>, <a href="/search/cs?searchtype=author&query=Fu%2C+X">Xingyu Fu</a>, <a href="/search/cs?searchtype=author&query=Roth%2C+D">Dan Roth</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Zettlemoyer%2C+L">Luke Zettlemoyer</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A Smith</a>, <a href="/search/cs?searchtype=author&query=Krishna%2C+R">Ranjay Krishna</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="2406.09403v3-abstract-short" style="display: inline;"> Humans draw to facilitate reasoning: we draw auxiliary lines when solving geometry problems; we mark and circle when reasoning on maps; we use sketches to amplify our ideas and relieve our limited-capacity working memory. However, such actions are missing in current multimodal language models (LMs). Current chain-of-thought and tool-use paradigms only use text as intermediate reasoning steps. In t… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.09403v3-abstract-full').style.display = 'inline'; document.getElementById('2406.09403v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2406.09403v3-abstract-full" style="display: none;"> Humans draw to facilitate reasoning: we draw auxiliary lines when solving geometry problems; we mark and circle when reasoning on maps; we use sketches to amplify our ideas and relieve our limited-capacity working memory. However, such actions are missing in current multimodal language models (LMs). Current chain-of-thought and tool-use paradigms only use text as intermediate reasoning steps. In this work, we introduce Sketchpad, a framework that gives multimodal LMs a visual sketchpad and tools to draw on the sketchpad. The LM conducts planning and reasoning according to the visual artifacts it has drawn. Different from prior work, which uses text-to-image models to enable LMs to draw, Sketchpad enables LMs to draw with lines, boxes, marks, etc., which is closer to human sketching and better facilitates reasoning. Sketchpad can also use specialist vision models during the sketching process (e.g., draw bounding boxes with object detection models, draw masks with segmentation models), to further enhance visual perception and reasoning. We experiment with a wide range of math tasks (including geometry, functions, graphs, and chess) and complex visual reasoning tasks. Sketchpad substantially improves performance on all tasks over strong base models with no sketching, yielding an average gain of 12.7% on math tasks, and 8.6% on vision tasks. GPT-4o with Sketchpad sets a new state of the art on all tasks, including V*Bench (80.3%), BLINK spatial reasoning (83.9%), and visual correspondence (80.8%). All codes and data are in https://visualsketchpad.github.io/. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.09403v3-abstract-full').style.display = 'none'; document.getElementById('2406.09403v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 10 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 13 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">Accepted to NeurIPS 2024. Project and codes url: https://visualsketchpad.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/2403.13112">arXiv:2403.13112</a> <span> [<a href="https://arxiv.org/pdf/2403.13112">pdf</a>, <a href="https://arxiv.org/format/2403.13112">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"> Efficient Encoder-Decoder Transformer Decoding for Decomposable Tasks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lu%2C+B">Bo-Ru Lu</a>, <a href="/search/cs?searchtype=author&query=Haduong%2C+N">Nikita Haduong</a>, <a href="/search/cs?searchtype=author&query=Lin%2C+C">Chien-Yu Lin</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+H">Hao Cheng</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A. Smith</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2403.13112v3-abstract-short" style="display: inline;"> Transformer-based NLP models are powerful but have high computational costs that limit deployment. Finetuned encoder-decoder models are popular in specialized domains and can outperform larger more generalized decoder-only models, such as GPT-4. We introduce a new configuration for encoder-decoder models that improves efficiency on structured output and decomposable tasks where multiple outputs ar… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.13112v3-abstract-full').style.display = 'inline'; document.getElementById('2403.13112v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2403.13112v3-abstract-full" style="display: none;"> Transformer-based NLP models are powerful but have high computational costs that limit deployment. Finetuned encoder-decoder models are popular in specialized domains and can outperform larger more generalized decoder-only models, such as GPT-4. We introduce a new configuration for encoder-decoder models that improves efficiency on structured output and decomposable tasks where multiple outputs are required for a single shared input. Our method, prompt-in-decoder (PiD), encodes the input once and decodes the output in parallel, boosting both training and inference efficiency by avoiding duplicate input encoding and increasing the operational intensity (ratio of numbers of arithmetic operation to memory access) of decoding process by sharing the input key-value cache. We achieve computation reduction that roughly scales with the number of subtasks, gaining up to 4.6x speed-up over state-of-the-art models for dialogue state tracking, summarization, and question-answering tasks, with comparable or better performance. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.13112v3-abstract-full').style.display = 'none'; document.getElementById('2403.13112v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 16 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 19 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 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">18 pages</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2311.09758">arXiv:2311.09758</a> <span> [<a href="https://arxiv.org/pdf/2311.09758">pdf</a>, <a href="https://arxiv.org/format/2311.09758">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"> OrchestraLLM: Efficient Orchestration of Language Models for Dialogue State Tracking </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lee%2C+C">Chia-Hsuan Lee</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+H">Hao Cheng</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2311.09758v3-abstract-short" style="display: inline;"> Large language models (LLMs) have revolutionized the landscape of Natural Language Processing systems, but are computationally expensive. To reduce the cost without sacrificing performance, previous studies have explored various approaches to harness the potential of Small Language Models (SLMs) as cost-effective alternatives to their larger counterparts. Driven by findings that SLMs and LLMs exhi… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.09758v3-abstract-full').style.display = 'inline'; document.getElementById('2311.09758v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2311.09758v3-abstract-full" style="display: none;"> Large language models (LLMs) have revolutionized the landscape of Natural Language Processing systems, but are computationally expensive. To reduce the cost without sacrificing performance, previous studies have explored various approaches to harness the potential of Small Language Models (SLMs) as cost-effective alternatives to their larger counterparts. Driven by findings that SLMs and LLMs exhibit complementary strengths in a structured knowledge extraction task, this work presents a novel SLM/LLM routing framework designed to improve computational efficiency and enhance task performance. First, exemplar pools are created to represent the types of contexts where each LM provides a more reliable answer, leveraging a sentence embedding fine-tuned so that context similarity is close to dialogue state similarity. Then, during inference, the k-nearest exemplars to the testing instance are retrieved, and the instance is routed according to majority vote. In dialogue state tracking tasks, the proposed routing framework enhances performance substantially compared to relying solely on LLMs, while reducing the computational costs by over 50%. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.09758v3-abstract-full').style.display = 'none'; document.getElementById('2311.09758v3-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, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 16 November, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 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">updated version (NAACL camera ready)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2307.07047">arXiv:2307.07047</a> <span> [<a href="https://arxiv.org/pdf/2307.07047">pdf</a>, <a href="https://arxiv.org/format/2307.07047">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"> Does Collaborative Human-LM Dialogue Generation Help Information Extraction from Human Dialogues? </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lu%2C+B">Bo-Ru Lu</a>, <a href="/search/cs?searchtype=author&query=Haduong%2C+N">Nikita Haduong</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+C">Chia-Hsuan Lee</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+Z">Zeqiu Wu</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+H">Hao Cheng</a>, <a href="/search/cs?searchtype=author&query=Koester%2C+P">Paul Koester</a>, <a href="/search/cs?searchtype=author&query=Utke%2C+J">Jean Utke</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+T">Tao Yu</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A. Smith</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2307.07047v2-abstract-short" style="display: inline;"> The capabilities of pretrained language models have opened opportunities to explore new application areas, but applications involving human-human interaction are limited by the fact that most data is protected from public release for privacy reasons. Problem-solving human dialogues in real applications can be much more complex than existing Wizard-of-Oz collections, preventing successful domain tr… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.07047v2-abstract-full').style.display = 'inline'; document.getElementById('2307.07047v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2307.07047v2-abstract-full" style="display: none;"> The capabilities of pretrained language models have opened opportunities to explore new application areas, but applications involving human-human interaction are limited by the fact that most data is protected from public release for privacy reasons. Problem-solving human dialogues in real applications can be much more complex than existing Wizard-of-Oz collections, preventing successful domain transfer. To support information extraction (IE) for a private call center dataset, we introduce a human-in-the-loop dialogue generation framework capable of synthesizing realistic dialogues. In IE experiments with auto insurance call center dialogues, we observe 25\% relative improvement in $F_1$ after augmenting a small set of real human conversations with synthetic data. We release code and our synthetic dataset to illustrate the complexity of real-world call center conversations and encourage development of complex dialogue datasets that are more representative of natural data. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.07047v2-abstract-full').style.display = 'none'; document.getElementById('2307.07047v2-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">v1</span> submitted 13 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/2306.09544">arXiv:2306.09544</a> <span> [<a href="https://arxiv.org/pdf/2306.09544">pdf</a>, <a href="https://arxiv.org/format/2306.09544">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"> Building blocks for complex tasks: Robust generative event extraction for radiology reports under domain shifts </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhou%2C+S">Sitong Zhou</a>, <a href="/search/cs?searchtype=author&query=Yetisgen%2C+M">Meliha Yetisgen</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2306.09544v1-abstract-short" style="display: inline;"> This paper explores methods for extracting information from radiology reports that generalize across exam modalities to reduce requirements for annotated data. We demonstrate that multi-pass T5-based text-to-text generative models exhibit better generalization across exam modalities compared to approaches that employ BERT-based task-specific classification layers. We then develop methods that redu… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.09544v1-abstract-full').style.display = 'inline'; document.getElementById('2306.09544v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2306.09544v1-abstract-full" style="display: none;"> This paper explores methods for extracting information from radiology reports that generalize across exam modalities to reduce requirements for annotated data. We demonstrate that multi-pass T5-based text-to-text generative models exhibit better generalization across exam modalities compared to approaches that employ BERT-based task-specific classification layers. We then develop methods that reduce the inference cost of the model, making large-scale corpus processing more feasible for clinical applications. Specifically, we introduce a generative technique that decomposes complex tasks into smaller subtask blocks, which improves a single-pass model when combined with multitask training. In addition, we leverage target-domain contexts during inference to enhance domain adaptation, enabling use of smaller models. Analyses offer insights into the benefits of different cost reduction strategies. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.09544v1-abstract-full').style.display = 'none'; document.getElementById('2306.09544v1-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 June, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> The 5th Clinical Natural Language Processing Workshop. At ACL 2023 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2306.01693">arXiv:2306.01693</a> <span> [<a href="https://arxiv.org/pdf/2306.01693">pdf</a>, <a href="https://arxiv.org/format/2306.01693">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"> Fine-Grained Human Feedback Gives Better Rewards for Language Model Training </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wu%2C+Z">Zeqiu Wu</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yushi Hu</a>, <a href="/search/cs?searchtype=author&query=Shi%2C+W">Weijia Shi</a>, <a href="/search/cs?searchtype=author&query=Dziri%2C+N">Nouha Dziri</a>, <a href="/search/cs?searchtype=author&query=Suhr%2C+A">Alane Suhr</a>, <a href="/search/cs?searchtype=author&query=Ammanabrolu%2C+P">Prithviraj Ammanabrolu</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A. Smith</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Hajishirzi%2C+H">Hannaneh Hajishirzi</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="2306.01693v2-abstract-short" style="display: inline;"> Language models (LMs) often exhibit undesirable text generation behaviors, including generating false, toxic, or irrelevant outputs. Reinforcement learning from human feedback (RLHF) - where human preference judgments on LM outputs are transformed into a learning signal - has recently shown promise in addressing these issues. However, such holistic feedback conveys limited information on long text… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.01693v2-abstract-full').style.display = 'inline'; document.getElementById('2306.01693v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2306.01693v2-abstract-full" style="display: none;"> Language models (LMs) often exhibit undesirable text generation behaviors, including generating false, toxic, or irrelevant outputs. Reinforcement learning from human feedback (RLHF) - where human preference judgments on LM outputs are transformed into a learning signal - has recently shown promise in addressing these issues. However, such holistic feedback conveys limited information on long text outputs; it does not indicate which aspects of the outputs influenced user preference; e.g., which parts contain what type(s) of errors. In this paper, we use fine-grained human feedback (e.g., which sentence is false, which sub-sentence is irrelevant) as an explicit training signal. We introduce Fine-Grained RLHF, a framework that enables training and learning from reward functions that are fine-grained in two respects: (1) density, providing a reward after every segment (e.g., a sentence) is generated; and (2) incorporating multiple reward models associated with different feedback types (e.g., factual incorrectness, irrelevance, and information incompleteness). We conduct experiments on detoxification and long-form question answering to illustrate how learning with such reward functions leads to improved performance, supported by both automatic and human evaluation. Additionally, we show that LM behaviors can be customized using different combinations of fine-grained reward models. We release all data, collected human feedback, and codes at https://FineGrainedRLHF.github.io. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.01693v2-abstract-full').style.display = 'none'; document.getElementById('2306.01693v2-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 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 2 June, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 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">NeurIPS 2023 camera-ready</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2303.11897">arXiv:2303.11897</a> <span> [<a href="https://arxiv.org/pdf/2303.11897">pdf</a>, <a href="https://arxiv.org/format/2303.11897">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"> TIFA: Accurate and Interpretable Text-to-Image Faithfulness Evaluation with Question Answering </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yushi Hu</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+B">Benlin Liu</a>, <a href="/search/cs?searchtype=author&query=Kasai%2C+J">Jungo Kasai</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yizhong Wang</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Krishna%2C+R">Ranjay Krishna</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A Smith</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="2303.11897v3-abstract-short" style="display: inline;"> Despite thousands of researchers, engineers, and artists actively working on improving text-to-image generation models, systems often fail to produce images that accurately align with the text inputs. We introduce TIFA (Text-to-Image Faithfulness evaluation with question Answering), an automatic evaluation metric that measures the faithfulness of a generated image to its text input via visual ques… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.11897v3-abstract-full').style.display = 'inline'; document.getElementById('2303.11897v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2303.11897v3-abstract-full" style="display: none;"> Despite thousands of researchers, engineers, and artists actively working on improving text-to-image generation models, systems often fail to produce images that accurately align with the text inputs. We introduce TIFA (Text-to-Image Faithfulness evaluation with question Answering), an automatic evaluation metric that measures the faithfulness of a generated image to its text input via visual question answering (VQA). Specifically, given a text input, we automatically generate several question-answer pairs using a language model. We calculate image faithfulness by checking whether existing VQA models can answer these questions using the generated image. TIFA is a reference-free metric that allows for fine-grained and interpretable evaluations of generated images. TIFA also has better correlations with human judgments than existing metrics. Based on this approach, we introduce TIFA v1.0, a benchmark consisting of 4K diverse text inputs and 25K questions across 12 categories (object, counting, etc.). We present a comprehensive evaluation of existing text-to-image models using TIFA v1.0 and highlight the limitations and challenges of current models. For instance, we find that current text-to-image models, despite doing well on color and material, still struggle in counting, spatial relations, and composing multiple objects. We hope our benchmark will help carefully measure the research progress in text-to-image synthesis and provide valuable insights for further research. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.11897v3-abstract-full').style.display = 'none'; document.getElementById('2303.11897v3-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 August, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 21 March, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 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">Accepted to 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/2212.09741">arXiv:2212.09741</a> <span> [<a href="https://arxiv.org/pdf/2212.09741">pdf</a>, <a href="https://arxiv.org/format/2212.09741">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"> One Embedder, Any Task: Instruction-Finetuned Text Embeddings </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Su%2C+H">Hongjin Su</a>, <a href="/search/cs?searchtype=author&query=Shi%2C+W">Weijia Shi</a>, <a href="/search/cs?searchtype=author&query=Kasai%2C+J">Jungo Kasai</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+Y">Yizhong Wang</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yushi Hu</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Yih%2C+W">Wen-tau Yih</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A. Smith</a>, <a href="/search/cs?searchtype=author&query=Zettlemoyer%2C+L">Luke Zettlemoyer</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+T">Tao Yu</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="2212.09741v3-abstract-short" style="display: inline;"> We introduce INSTRUCTOR, a new method for computing text embeddings given task instructions: every text input is embedded together with instructions explaining the use case (e.g., task and domain descriptions). Unlike encoders from prior work that are more specialized, INSTRUCTOR is a single embedder that can generate text embeddings tailored to different downstream tasks and domains, without any… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2212.09741v3-abstract-full').style.display = 'inline'; document.getElementById('2212.09741v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2212.09741v3-abstract-full" style="display: none;"> We introduce INSTRUCTOR, a new method for computing text embeddings given task instructions: every text input is embedded together with instructions explaining the use case (e.g., task and domain descriptions). Unlike encoders from prior work that are more specialized, INSTRUCTOR is a single embedder that can generate text embeddings tailored to different downstream tasks and domains, without any further training. We first annotate instructions for 330 diverse tasks and train INSTRUCTOR on this multitask mixture with a contrastive loss. We evaluate INSTRUCTOR on 70 embedding evaluation tasks (66 of which are unseen during training), ranging from classification and information retrieval to semantic textual similarity and text generation evaluation. INSTRUCTOR, while having an order of magnitude fewer parameters than the previous best model, achieves state-of-the-art performance, with an average improvement of 3.4% compared to the previous best results on the 70 diverse datasets. Our analysis suggests that INSTRUCTOR is robust to changes in instructions, and that instruction finetuning mitigates the challenge of training a single model on diverse datasets. Our model, code, and data are available at https://instructor-embedding.github.io. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2212.09741v3-abstract-full').style.display = 'none'; document.getElementById('2212.09741v3-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, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 19 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">Accepted in ACL2023 Findings</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2210.02875">arXiv:2210.02875</a> <span> [<a href="https://arxiv.org/pdf/2210.02875">pdf</a>, <a href="https://arxiv.org/format/2210.02875">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"> Binding Language Models in Symbolic Languages </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Cheng%2C+Z">Zhoujun Cheng</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+T">Tianbao Xie</a>, <a href="/search/cs?searchtype=author&query=Shi%2C+P">Peng Shi</a>, <a href="/search/cs?searchtype=author&query=Li%2C+C">Chengzu Li</a>, <a href="/search/cs?searchtype=author&query=Nadkarni%2C+R">Rahul Nadkarni</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yushi Hu</a>, <a href="/search/cs?searchtype=author&query=Xiong%2C+C">Caiming Xiong</a>, <a href="/search/cs?searchtype=author&query=Radev%2C+D">Dragomir Radev</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Zettlemoyer%2C+L">Luke Zettlemoyer</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A. Smith</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+T">Tao Yu</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="2210.02875v2-abstract-short" style="display: inline;"> Though end-to-end neural approaches have recently been dominating NLP tasks in both performance and ease-of-use, they lack interpretability and robustness. We propose Binder, a training-free neural-symbolic framework that maps the task input to a program, which (1) allows binding a unified API of language model (LM) functionalities to a programming language (e.g., SQL, Python) to extend its gramma… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2210.02875v2-abstract-full').style.display = 'inline'; document.getElementById('2210.02875v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2210.02875v2-abstract-full" style="display: none;"> Though end-to-end neural approaches have recently been dominating NLP tasks in both performance and ease-of-use, they lack interpretability and robustness. We propose Binder, a training-free neural-symbolic framework that maps the task input to a program, which (1) allows binding a unified API of language model (LM) functionalities to a programming language (e.g., SQL, Python) to extend its grammar coverage and thus tackle more diverse questions, (2) adopts an LM as both the program parser and the underlying model called by the API during execution, and (3) requires only a few in-context exemplar annotations. Specifically, we employ GPT-3 Codex as the LM. In the parsing stage, with only a few in-context exemplars, Codex is able to identify the part of the task input that cannot be answerable by the original programming language, correctly generate API calls to prompt Codex to solve the unanswerable part, and identify where to place the API calls while being compatible with the original grammar. In the execution stage, Codex can perform versatile functionalities (e.g., commonsense QA, information extraction) given proper prompts in the API calls. Binder achieves state-of-the-art results on WikiTableQuestions and TabFact datasets, with explicit output programs that benefit human debugging. Note that previous best systems are all finetuned on tens of thousands of task-specific samples, while Binder only uses dozens of annotations as in-context exemplars without any training. Our code is available at https://github.com/HKUNLP/Binder . <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2210.02875v2-abstract-full').style.display = 'none'; document.getElementById('2210.02875v2-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 February, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 6 October, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 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">ICLR 2023 camera ready, 27 pages, 10 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/2209.09485">arXiv:2209.09485</a> <span> [<a href="https://arxiv.org/pdf/2209.09485">pdf</a>, <a href="https://arxiv.org/format/2209.09485">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"> Generalizing through Forgetting -- Domain Generalization for Symptom Event Extraction in Clinical Notes </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhou%2C+S">Sitong Zhou</a>, <a href="/search/cs?searchtype=author&query=Lybarger%2C+K">Kevin Lybarger</a>, <a href="/search/cs?searchtype=author&query=Yetisgen%2C+M">Meliha Yetisgen</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2209.09485v2-abstract-short" style="display: inline;"> Symptom information is primarily documented in free-text clinical notes and is not directly accessible for downstream applications. To address this challenge, information extraction approaches that can handle clinical language variation across different institutions and specialties are needed. In this paper, we present domain generalization for symptom extraction using pretraining and fine-tuning… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2209.09485v2-abstract-full').style.display = 'inline'; document.getElementById('2209.09485v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2209.09485v2-abstract-full" style="display: none;"> Symptom information is primarily documented in free-text clinical notes and is not directly accessible for downstream applications. To address this challenge, information extraction approaches that can handle clinical language variation across different institutions and specialties are needed. In this paper, we present domain generalization for symptom extraction using pretraining and fine-tuning data that differs from the target domain in terms of institution and/or specialty and patient population. We extract symptom events using a transformer-based joint entity and relation extraction method. To reduce reliance on domain-specific features, we propose a domain generalization method that dynamically masks frequent symptoms words in the source domain. Additionally, we pretrain the transformer language model (LM) on task-related unlabeled texts for better representation. Our experiments indicate that masking and adaptive pretraining methods can significantly improve performance when the source domain is more distant from the target domain. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2209.09485v2-abstract-full').style.display = 'none'; document.getElementById('2209.09485v2-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 February, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 20 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">Journal ref:</span> AMIA 2023 Informatics Summit </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2209.01975">arXiv:2209.01975</a> <span> [<a href="https://arxiv.org/pdf/2209.01975">pdf</a>, <a href="https://arxiv.org/format/2209.01975">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"> Selective Annotation Makes Language Models Better Few-Shot Learners </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Su%2C+H">Hongjin Su</a>, <a href="/search/cs?searchtype=author&query=Kasai%2C+J">Jungo Kasai</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+C+H">Chen Henry Wu</a>, <a href="/search/cs?searchtype=author&query=Shi%2C+W">Weijia Shi</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+T">Tianlu Wang</a>, <a href="/search/cs?searchtype=author&query=Xin%2C+J">Jiayi Xin</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+R">Rui Zhang</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Zettlemoyer%2C+L">Luke Zettlemoyer</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A. Smith</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+T">Tao Yu</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="2209.01975v1-abstract-short" style="display: inline;"> Many recent approaches to natural language tasks are built on the remarkable abilities of large language models. Large language models can perform in-context learning, where they learn a new task from a few task demonstrations, without any parameter updates. This work examines the implications of in-context learning for the creation of datasets for new natural language tasks. Departing from recent… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2209.01975v1-abstract-full').style.display = 'inline'; document.getElementById('2209.01975v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2209.01975v1-abstract-full" style="display: none;"> Many recent approaches to natural language tasks are built on the remarkable abilities of large language models. Large language models can perform in-context learning, where they learn a new task from a few task demonstrations, without any parameter updates. This work examines the implications of in-context learning for the creation of datasets for new natural language tasks. Departing from recent in-context learning methods, we formulate an annotation-efficient, two-step framework: selective annotation that chooses a pool of examples to annotate from unlabeled data in advance, followed by prompt retrieval that retrieves task examples from the annotated pool at test time. Based on this framework, we propose an unsupervised, graph-based selective annotation method, voke-k, to select diverse, representative examples to annotate. Extensive experiments on 10 datasets (covering classification, commonsense reasoning, dialogue, and text/code generation) demonstrate that our selective annotation method improves the task performance by a large margin. On average, vote-k achieves a 12.9%/11.4% relative gain under an annotation budget of 18/100, as compared to randomly selecting examples to annotate. Compared to state-of-the-art supervised finetuning approaches, it yields similar performance with 10-100x less annotation cost across 10 tasks. We further analyze the effectiveness of our framework in various scenarios: language models with varying sizes, alternative selective annotation methods, and cases where there is a test data domain shift. We hope that our studies will serve as a basis for data annotations as large language models are increasingly applied to new tasks. Our code is available at https://github.com/HKUNLP/icl-selective-annotation. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2209.01975v1-abstract-full').style.display = 'none'; document.getElementById('2209.01975v1-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 September, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2022. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2207.00746">arXiv:2207.00746</a> <span> [<a href="https://arxiv.org/pdf/2207.00746">pdf</a>, <a href="https://arxiv.org/format/2207.00746">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"> INSCIT: Information-Seeking Conversations with Mixed-Initiative Interactions </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wu%2C+Z">Zeqiu Wu</a>, <a href="/search/cs?searchtype=author&query=Parish%2C+R">Ryu Parish</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+H">Hao Cheng</a>, <a href="/search/cs?searchtype=author&query=Min%2C+S">Sewon Min</a>, <a href="/search/cs?searchtype=author&query=Ammanabrolu%2C+P">Prithviraj Ammanabrolu</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Hajishirzi%2C+H">Hannaneh Hajishirzi</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="2207.00746v2-abstract-short" style="display: inline;"> In an information-seeking conversation, a user may ask questions that are under-specified or unanswerable. An ideal agent would interact by initiating different response types according to the available knowledge sources. However, most current studies either fail to or artificially incorporate such agent-side initiative. This work presents InSCIt, a dataset for Information-Seeking Conversations wi… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2207.00746v2-abstract-full').style.display = 'inline'; document.getElementById('2207.00746v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2207.00746v2-abstract-full" style="display: none;"> In an information-seeking conversation, a user may ask questions that are under-specified or unanswerable. An ideal agent would interact by initiating different response types according to the available knowledge sources. However, most current studies either fail to or artificially incorporate such agent-side initiative. This work presents InSCIt, a dataset for Information-Seeking Conversations with mixed-initiative Interactions. It contains 4.7K user-agent turns from 805 human-human conversations where the agent searches over Wikipedia and either directly answers, asks for clarification, or provides relevant information to address user queries. The data supports two subtasks, evidence passage identification and response generation, as well as a human evaluation protocol to assess model performance. We report results of two systems based on state-of-the-art models of conversational knowledge identification and open-domain question answering. Both systems significantly underperform humans, suggesting ample room for improvement in future studies. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2207.00746v2-abstract-full').style.display = 'none'; document.getElementById('2207.00746v2-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 June, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 2 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">TACL 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/2205.12244">arXiv:2205.12244</a> <span> [<a href="https://arxiv.org/pdf/2205.12244">pdf</a>, <a href="https://arxiv.org/format/2205.12244">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"> Unsupervised Learning of Hierarchical Conversation Structure </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lu%2C+B">Bo-Ru Lu</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yushi Hu</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+H">Hao Cheng</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A. Smith</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2205.12244v2-abstract-short" style="display: inline;"> Human conversations can evolve in many different ways, creating challenges for automatic understanding and summarization. Goal-oriented conversations often have meaningful sub-dialogue structure, but it can be highly domain-dependent. This work introduces an unsupervised approach to learning hierarchical conversation structure, including turn and sub-dialogue segment labels, corresponding roughly… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2205.12244v2-abstract-full').style.display = 'inline'; document.getElementById('2205.12244v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2205.12244v2-abstract-full" style="display: none;"> Human conversations can evolve in many different ways, creating challenges for automatic understanding and summarization. Goal-oriented conversations often have meaningful sub-dialogue structure, but it can be highly domain-dependent. This work introduces an unsupervised approach to learning hierarchical conversation structure, including turn and sub-dialogue segment labels, corresponding roughly to dialogue acts and sub-tasks, respectively. The decoded structure is shown to be useful in enhancing neural models of language for three conversation-level understanding tasks. Further, the learned finite-state sub-dialogue network is made interpretable through automatic summarization. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2205.12244v2-abstract-full').style.display = 'none'; document.getElementById('2205.12244v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 17 November, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 24 May, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 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">In Findings of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2022 Findings)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2204.00967">arXiv:2204.00967</a> <span> [<a href="https://arxiv.org/pdf/2204.00967">pdf</a>, <a href="https://arxiv.org/format/2204.00967">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Audio and Speech Processing">eess.AS</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Sound">cs.SD</span> </div> </div> <p class="title is-5 mathjax"> Automatic Dialect Density Estimation for African American English </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Johnson%2C+A">Alexander Johnson</a>, <a href="/search/cs?searchtype=author&query=Everson%2C+K">Kevin Everson</a>, <a href="/search/cs?searchtype=author&query=Ravi%2C+V">Vijay Ravi</a>, <a href="/search/cs?searchtype=author&query=Gladney%2C+A">Anissa Gladney</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Alwan%2C+A">Abeer Alwan</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="2204.00967v1-abstract-short" style="display: inline;"> In this paper, we explore automatic prediction of dialect density of the African American English (AAE) dialect, where dialect density is defined as the percentage of words in an utterance that contain characteristics of the non-standard dialect. We investigate several acoustic and language modeling features, including the commonly used X-vector representation and ComParE feature set, in addition… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2204.00967v1-abstract-full').style.display = 'inline'; document.getElementById('2204.00967v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2204.00967v1-abstract-full" style="display: none;"> In this paper, we explore automatic prediction of dialect density of the African American English (AAE) dialect, where dialect density is defined as the percentage of words in an utterance that contain characteristics of the non-standard dialect. We investigate several acoustic and language modeling features, including the commonly used X-vector representation and ComParE feature set, in addition to information extracted from ASR transcripts of the audio files and prosodic information. To address issues of limited labeled data, we use a weakly supervised model to project prosodic and X-vector features into low-dimensional task-relevant representations. An XGBoost model is then used to predict the speaker's dialect density from these features and show which are most significant during inference. We evaluate the utility of these features both alone and in combination for the given task. This work, which does not rely on hand-labeled transcripts, is performed on audio segments from the CORAAL database. We show a significant correlation between our predicted and ground truth dialect density measures for AAE speech in this database and propose this work as a tool for explaining and mitigating bias in speech technology. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2204.00967v1-abstract-full').style.display = 'none'; document.getElementById('2204.00967v1-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 April, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 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">5 pages, 2 figures</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> F.2.2; I.2.7 </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> Interspeech 2022 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2203.08568">arXiv:2203.08568</a> <span> [<a href="https://arxiv.org/pdf/2203.08568">pdf</a>, <a href="https://arxiv.org/format/2203.08568">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"> In-Context Learning for Few-Shot Dialogue State Tracking </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yushi Hu</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+C">Chia-Hsuan Lee</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+T">Tianbao Xie</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+T">Tao Yu</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A. Smith</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2203.08568v3-abstract-short" style="display: inline;"> Collecting and annotating task-oriented dialogues is time-consuming and costly; thus, zero and few shot learning could greatly benefit dialogue state tracking (DST). In this work, we propose an in-context learning (ICL) framework for zero-shot and few-shot learning DST, where a large pre-trained language model (LM) takes a test instance and a few exemplars as input, and directly decodes the dialog… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2203.08568v3-abstract-full').style.display = 'inline'; document.getElementById('2203.08568v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2203.08568v3-abstract-full" style="display: none;"> Collecting and annotating task-oriented dialogues is time-consuming and costly; thus, zero and few shot learning could greatly benefit dialogue state tracking (DST). In this work, we propose an in-context learning (ICL) framework for zero-shot and few-shot learning DST, where a large pre-trained language model (LM) takes a test instance and a few exemplars as input, and directly decodes the dialogue state without any parameter updates. To better leverage a tabular domain description in the LM prompt, we reformulate DST into a text-to-SQL problem. We also propose a novel approach to retrieve annotated dialogues as exemplars. Empirical results on MultiWOZ show that our method IC-DST substantially outperforms previous fine-tuned state-of-the-art models in few-shot settings. In addition, we test IC-DST in zero-shot settings, in which the model only takes a fixed task instruction as input, finding that it outperforms previous zero-shot methods by a large margin. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2203.08568v3-abstract-full').style.display = 'none'; document.getElementById('2203.08568v3-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 October, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 16 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">To appear in Findings of EMNLP 2022</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2112.08558">arXiv:2112.08558</a> <span> [<a href="https://arxiv.org/pdf/2112.08558">pdf</a>, <a href="https://arxiv.org/format/2112.08558">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"> CONQRR: Conversational Query Rewriting for Retrieval with Reinforcement Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wu%2C+Z">Zeqiu Wu</a>, <a href="/search/cs?searchtype=author&query=Luan%2C+Y">Yi Luan</a>, <a href="/search/cs?searchtype=author&query=Rashkin%2C+H">Hannah Rashkin</a>, <a href="/search/cs?searchtype=author&query=Reitter%2C+D">David Reitter</a>, <a href="/search/cs?searchtype=author&query=Hajishirzi%2C+H">Hannaneh Hajishirzi</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Tomar%2C+G+S">Gaurav Singh Tomar</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="2112.08558v3-abstract-short" style="display: inline;"> Compared to standard retrieval tasks, passage retrieval for conversational question answering (CQA) poses new challenges in understanding the current user question, as each question needs to be interpreted within the dialogue context. Moreover, it can be expensive to re-train well-established retrievers such as search engines that are originally developed for non-conversational queries. To facilit… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2112.08558v3-abstract-full').style.display = 'inline'; document.getElementById('2112.08558v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2112.08558v3-abstract-full" style="display: none;"> Compared to standard retrieval tasks, passage retrieval for conversational question answering (CQA) poses new challenges in understanding the current user question, as each question needs to be interpreted within the dialogue context. Moreover, it can be expensive to re-train well-established retrievers such as search engines that are originally developed for non-conversational queries. To facilitate their use, we develop a query rewriting model CONQRR that rewrites a conversational question in the context into a standalone question. It is trained with a novel reward function to directly optimize towards retrieval using reinforcement learning and can be adapted to any off-the-shelf retriever. CONQRR achieves state-of-the-art results on a recent open-domain CQA dataset containing conversations from three different sources, and is effective for two different off-the-shelf retrievers. Our extensive analysis also shows the robustness of CONQRR to out-of-domain dialogues as well as to zero query rewriting supervision. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2112.08558v3-abstract-full').style.display = 'none'; document.getElementById('2112.08558v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 28 October, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 15 December, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 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">EMNLP 2022 camera-ready</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2109.07506">arXiv:2109.07506</a> <span> [<a href="https://arxiv.org/pdf/2109.07506">pdf</a>, <a href="https://arxiv.org/format/2109.07506">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"> Dialogue State Tracking with a Language Model using Schema-Driven Prompting </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lee%2C+C">Chia-Hsuan Lee</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+H">Hao Cheng</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2109.07506v1-abstract-short" style="display: inline;"> Task-oriented conversational systems often use dialogue state tracking to represent the user's intentions, which involves filling in values of pre-defined slots. Many approaches have been proposed, often using task-specific architectures with special-purpose classifiers. Recently, good results have been obtained using more general architectures based on pretrained language models. Here, we introdu… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2109.07506v1-abstract-full').style.display = 'inline'; document.getElementById('2109.07506v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2109.07506v1-abstract-full" style="display: none;"> Task-oriented conversational systems often use dialogue state tracking to represent the user's intentions, which involves filling in values of pre-defined slots. Many approaches have been proposed, often using task-specific architectures with special-purpose classifiers. Recently, good results have been obtained using more general architectures based on pretrained language models. Here, we introduce a new variation of the language modeling approach that uses schema-driven prompting to provide task-aware history encoding that is used for both categorical and non-categorical slots. We further improve performance by augmenting the prompting with schema descriptions, a naturally occurring source of in-domain knowledge. Our purely generative system achieves state-of-the-art performance on MultiWOZ 2.2 and achieves competitive performance on two other benchmarks: MultiWOZ 2.1 and M2M. The data and code will be available at https://github.com/chiahsuan156/DST-as-Prompting. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2109.07506v1-abstract-full').style.display = 'none'; document.getElementById('2109.07506v1-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 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">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted to EMNLP 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/2109.04673">arXiv:2109.04673</a> <span> [<a href="https://arxiv.org/pdf/2109.04673">pdf</a>, <a href="https://arxiv.org/format/2109.04673">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"> DIALKI: Knowledge Identification in Conversational Systems through Dialogue-Document Contextualization </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wu%2C+Z">Zeqiu Wu</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+B">Bo-Ru Lu</a>, <a href="/search/cs?searchtype=author&query=Hajishirzi%2C+H">Hannaneh Hajishirzi</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2109.04673v1-abstract-short" style="display: inline;"> Identifying relevant knowledge to be used in conversational systems that are grounded in long documents is critical to effective response generation. We introduce a knowledge identification model that leverages the document structure to provide dialogue-contextualized passage encodings and better locate knowledge relevant to the conversation. An auxiliary loss captures the history of dialogue-docu… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2109.04673v1-abstract-full').style.display = 'inline'; document.getElementById('2109.04673v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2109.04673v1-abstract-full" style="display: none;"> Identifying relevant knowledge to be used in conversational systems that are grounded in long documents is critical to effective response generation. We introduce a knowledge identification model that leverages the document structure to provide dialogue-contextualized passage encodings and better locate knowledge relevant to the conversation. An auxiliary loss captures the history of dialogue-document connections. We demonstrate the effectiveness of our model on two document-grounded conversational datasets and provide analyses showing generalization to unseen documents and long dialogue contexts. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2109.04673v1-abstract-full').style.display = 'none'; document.getElementById('2109.04673v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 10 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">Comments:</span> <span class="has-text-grey-dark mathjax">EMNLP 2021 camera-ready</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2106.07794">arXiv:2106.07794</a> <span> [<a href="https://arxiv.org/pdf/2106.07794">pdf</a>, <a href="https://arxiv.org/format/2106.07794">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"> Assessing the Use of Prosody in Constituency Parsing of Imperfect Transcripts </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Tran%2C+T">Trang Tran</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2106.07794v1-abstract-short" style="display: inline;"> This work explores constituency parsing on automatically recognized transcripts of conversational speech. The neural parser is based on a sentence encoder that leverages word vectors contextualized with prosodic features, jointly learning prosodic feature extraction with parsing. We assess the utility of the prosody in parsing on imperfect transcripts, i.e. transcripts with automatic speech recogn… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.07794v1-abstract-full').style.display = 'inline'; document.getElementById('2106.07794v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2106.07794v1-abstract-full" style="display: none;"> This work explores constituency parsing on automatically recognized transcripts of conversational speech. The neural parser is based on a sentence encoder that leverages word vectors contextualized with prosodic features, jointly learning prosodic feature extraction with parsing. We assess the utility of the prosody in parsing on imperfect transcripts, i.e. transcripts with automatic speech recognition (ASR) errors, by applying the parser in an N-best reranking framework. In experiments on Switchboard, we obtain 13-15% of the oracle N-best gain relative to parsing the 1-best ASR output, with insignificant impact on word recognition error rate. Prosody provides a significant part of the gain, and analyses suggest that it leads to more grammatical utterances via recovering function words. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.07794v1-abstract-full').style.display = 'none'; document.getElementById('2106.07794v1-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 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 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">Interspeech 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/2101.10573">arXiv:2101.10573</a> <span> [<a href="https://arxiv.org/pdf/2101.10573">pdf</a>, <a href="https://arxiv.org/format/2101.10573">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"> Representations for Question Answering from Documents with Tables and Text </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zayats%2C+V">Vicky Zayats</a>, <a href="/search/cs?searchtype=author&query=Toutanova%2C+K">Kristina Toutanova</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2101.10573v1-abstract-short" style="display: inline;"> Tables in Web documents are pervasive and can be directly used to answer many of the queries searched on the Web, motivating their integration in question answering. Very often information presented in tables is succinct and hard to interpret with standard language representations. On the other hand, tables often appear within textual context, such as an article describing the table. Using the inf… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2101.10573v1-abstract-full').style.display = 'inline'; document.getElementById('2101.10573v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2101.10573v1-abstract-full" style="display: none;"> Tables in Web documents are pervasive and can be directly used to answer many of the queries searched on the Web, motivating their integration in question answering. Very often information presented in tables is succinct and hard to interpret with standard language representations. On the other hand, tables often appear within textual context, such as an article describing the table. Using the information from an article as additional context can potentially enrich table representations. In this work we aim to improve question answering from tables by refining table representations based on information from surrounding text. We also present an effective method to combine text and table-based predictions for question answering from full documents, obtaining significant improvements on the Natural Questions dataset. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2101.10573v1-abstract-full').style.display = 'none'; document.getElementById('2101.10573v1-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 January, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 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">To appear at EACL 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/2012.00974">arXiv:2012.00974</a> <span> [<a href="https://arxiv.org/pdf/2012.00974">pdf</a>, <a href="https://arxiv.org/format/2012.00974">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="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Extracting COVID-19 Diagnoses and Symptoms From Clinical Text: A New Annotated Corpus and Neural Event Extraction Framework </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lybarger%2C+K">Kevin Lybarger</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Thompson%2C+M">Matthew Thompson</a>, <a href="/search/cs?searchtype=author&query=Yetisgen%2C+M">Meliha Yetisgen</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="2012.00974v2-abstract-short" style="display: inline;"> Coronavirus disease 2019 (COVID-19) is a global pandemic. Although much has been learned about the novel coronavirus since its emergence, there are many open questions related to tracking its spread, describing symptomology, predicting the severity of infection, and forecasting healthcare utilization. Free-text clinical notes contain critical information for resolving these questions. Data-driven,… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2012.00974v2-abstract-full').style.display = 'inline'; document.getElementById('2012.00974v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2012.00974v2-abstract-full" style="display: none;"> Coronavirus disease 2019 (COVID-19) is a global pandemic. Although much has been learned about the novel coronavirus since its emergence, there are many open questions related to tracking its spread, describing symptomology, predicting the severity of infection, and forecasting healthcare utilization. Free-text clinical notes contain critical information for resolving these questions. Data-driven, automatic information extraction models are needed to use this text-encoded information in large-scale studies. This work presents a new clinical corpus, referred to as the COVID-19 Annotated Clinical Text (CACT) Corpus, which comprises 1,472 notes with detailed annotations characterizing COVID-19 diagnoses, testing, and clinical presentation. We introduce a span-based event extraction model that jointly extracts all annotated phenomena, achieving high performance in identifying COVID-19 and symptom events with associated assertion values (0.83-0.97 F1 for events and 0.73-0.79 F1 for assertions). In a secondary use application, we explored the prediction of COVID-19 test results using structured patient data (e.g. vital signs and laboratory results) and automatically extracted symptom information. The automatically extracted symptoms improve prediction performance, beyond structured data alone. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2012.00974v2-abstract-full').style.display = 'none'; document.getElementById('2012.00974v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 10 March, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 2 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/2010.04293">arXiv:2010.04293</a> <span> [<a href="https://arxiv.org/pdf/2010.04293">pdf</a>, <a href="https://arxiv.org/format/2010.04293">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"> Analysis of Disfluency in Children's Speech </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Tran%2C+T">Trang Tran</a>, <a href="/search/cs?searchtype=author&query=Tinkler%2C+M">Morgan Tinkler</a>, <a href="/search/cs?searchtype=author&query=Yeung%2C+G">Gary Yeung</a>, <a href="/search/cs?searchtype=author&query=Alwan%2C+A">Abeer Alwan</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2010.04293v1-abstract-short" style="display: inline;"> Disfluencies are prevalent in spontaneous speech, as shown in many studies of adult speech. Less is understood about children's speech, especially in pre-school children who are still developing their language skills. We present a novel dataset with annotated disfluencies of spontaneous explanations from 26 children (ages 5--8), interviewed twice over a year-long period. Our preliminary analysis r… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2010.04293v1-abstract-full').style.display = 'inline'; document.getElementById('2010.04293v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2010.04293v1-abstract-full" style="display: none;"> Disfluencies are prevalent in spontaneous speech, as shown in many studies of adult speech. Less is understood about children's speech, especially in pre-school children who are still developing their language skills. We present a novel dataset with annotated disfluencies of spontaneous explanations from 26 children (ages 5--8), interviewed twice over a year-long period. Our preliminary analysis reveals significant differences between children's speech in our corpus and adult spontaneous speech from two corpora (Switchboard and CallHome). Children have higher disfluency and filler rates, tend to use nasal filled pauses more frequently, and on average exhibit longer reparandums than repairs, in contrast to adult speakers. Despite the differences, an automatic disfluency detection system trained on adult (Switchboard) speech transcripts performs reasonably well on children's speech, achieving an F1 score that is 10\% higher than the score on an adult out-of-domain dataset (CallHome). <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2010.04293v1-abstract-full').style.display = 'none'; document.getElementById('2010.04293v1-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, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2020. </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">Interspeech 2020</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2010.04288">arXiv:2010.04288</a> <span> [<a href="https://arxiv.org/pdf/2010.04288">pdf</a>, <a href="https://arxiv.org/format/2010.04288">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"> On the Role of Style in Parsing Speech with Neural Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Tran%2C+T">Trang Tran</a>, <a href="/search/cs?searchtype=author&query=Yuan%2C+J">Jiahong Yuan</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Y">Yang Liu</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="2010.04288v1-abstract-short" style="display: inline;"> The differences in written text and conversational speech are substantial; previous parsers trained on treebanked text have given very poor results on spontaneous speech. For spoken language, the mismatch in style also extends to prosodic cues, though it is less well understood. This paper re-examines the use of written text in parsing speech in the context of recent advances in neural language pr… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2010.04288v1-abstract-full').style.display = 'inline'; document.getElementById('2010.04288v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2010.04288v1-abstract-full" style="display: none;"> The differences in written text and conversational speech are substantial; previous parsers trained on treebanked text have given very poor results on spontaneous speech. For spoken language, the mismatch in style also extends to prosodic cues, though it is less well understood. This paper re-examines the use of written text in parsing speech in the context of recent advances in neural language processing. We show that neural approaches facilitate using written text to improve parsing of spontaneous speech, and that prosody further improves over this state-of-the-art result. Further, we find an asymmetric degradation from read vs. spontaneous mismatch, with spontaneous speech more generally useful for training parsers. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2010.04288v1-abstract-full').style.display = 'none'; document.getElementById('2010.04288v1-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, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2020. </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">Interspeech 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/2009.09162">arXiv:2009.09162</a> <span> [<a href="https://arxiv.org/pdf/2009.09162">pdf</a>, <a href="https://arxiv.org/format/2009.09162">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"> Extracting Summary Knowledge Graphs from Long Documents </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wu%2C+Z">Zeqiu Wu</a>, <a href="/search/cs?searchtype=author&query=Koncel-Kedziorski%2C+R">Rik Koncel-Kedziorski</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Hajishirzi%2C+H">Hannaneh Hajishirzi</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="2009.09162v2-abstract-short" style="display: inline;"> Knowledge graphs capture entities and relations from long documents and can facilitate reasoning in many downstream applications. Extracting compact knowledge graphs containing only salient entities and relations is important but challenging for understanding and summarizing long documents. We introduce a new text-to-graph task of predicting summarized knowledge graphs from long documents. We deve… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2009.09162v2-abstract-full').style.display = 'inline'; document.getElementById('2009.09162v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2009.09162v2-abstract-full" style="display: none;"> Knowledge graphs capture entities and relations from long documents and can facilitate reasoning in many downstream applications. Extracting compact knowledge graphs containing only salient entities and relations is important but challenging for understanding and summarizing long documents. We introduce a new text-to-graph task of predicting summarized knowledge graphs from long documents. We develop a dataset of 200k document/graph pairs using automatic and human annotations. We also develop strong baselines for this task based on graph learning and text summarization, and provide quantitative and qualitative studies of their effect. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2009.09162v2-abstract-full').style.display = 'none'; document.getElementById('2009.09162v2-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 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 19 September, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2020. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2005.00613">arXiv:2005.00613</a> <span> [<a href="https://arxiv.org/pdf/2005.00613">pdf</a>, <a href="https://arxiv.org/format/2005.00613">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"> A Controllable Model of Grounded Response Generation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wu%2C+Z">Zeqiu Wu</a>, <a href="/search/cs?searchtype=author&query=Galley%2C+M">Michel Galley</a>, <a href="/search/cs?searchtype=author&query=Brockett%2C+C">Chris Brockett</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Y">Yizhe Zhang</a>, <a href="/search/cs?searchtype=author&query=Gao%2C+X">Xiang Gao</a>, <a href="/search/cs?searchtype=author&query=Quirk%2C+C">Chris Quirk</a>, <a href="/search/cs?searchtype=author&query=Koncel-Kedziorski%2C+R">Rik Koncel-Kedziorski</a>, <a href="/search/cs?searchtype=author&query=Gao%2C+J">Jianfeng Gao</a>, <a href="/search/cs?searchtype=author&query=Hajishirzi%2C+H">Hannaneh Hajishirzi</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Dolan%2C+B">Bill Dolan</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="2005.00613v2-abstract-short" style="display: inline;"> Current end-to-end neural conversation models inherently lack the flexibility to impose semantic control in the response generation process, often resulting in uninteresting responses. Attempts to boost informativeness alone come at the expense of factual accuracy, as attested by pretrained language models' propensity to "hallucinate" facts. While this may be mitigated by access to background know… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2005.00613v2-abstract-full').style.display = 'inline'; document.getElementById('2005.00613v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2005.00613v2-abstract-full" style="display: none;"> Current end-to-end neural conversation models inherently lack the flexibility to impose semantic control in the response generation process, often resulting in uninteresting responses. Attempts to boost informativeness alone come at the expense of factual accuracy, as attested by pretrained language models' propensity to "hallucinate" facts. While this may be mitigated by access to background knowledge, there is scant guarantee of relevance and informativeness in generated responses. We propose a framework that we call controllable grounded response generation (CGRG), in which lexical control phrases are either provided by a user or automatically extracted by a control phrase predictor from dialogue context and grounding knowledge. Quantitative and qualitative results show that, using this framework, a transformer based model with a novel inductive attention mechanism, trained on a conversation-like Reddit dataset, outperforms strong generation baselines. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2005.00613v2-abstract-full').style.display = 'none'; document.getElementById('2005.00613v2-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 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 1 May, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2020. </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">AAAI 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/2004.05438">arXiv:2004.05438</a> <span> [<a href="https://arxiv.org/pdf/2004.05438">pdf</a>, <a href="https://arxiv.org/format/2004.05438">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="Machine Learning">cs.LG</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/10.1016/j.jbi.2020.103631">10.1016/j.jbi.2020.103631 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Annotating Social Determinants of Health Using Active Learning, and Characterizing Determinants Using Neural Event Extraction </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lybarger%2C+K">Kevin Lybarger</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Yetisgen%2C+M">Meliha Yetisgen</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="2004.05438v2-abstract-short" style="display: inline;"> Social determinants of health (SDOH) affect health outcomes, and knowledge of SDOH can inform clinical decision-making. Automatically extracting SDOH information from clinical text requires data-driven information extraction models trained on annotated corpora that are heterogeneous and frequently include critical SDOH. This work presents a new corpus with SDOH annotations, a novel active learning… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2004.05438v2-abstract-full').style.display = 'inline'; document.getElementById('2004.05438v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2004.05438v2-abstract-full" style="display: none;"> Social determinants of health (SDOH) affect health outcomes, and knowledge of SDOH can inform clinical decision-making. Automatically extracting SDOH information from clinical text requires data-driven information extraction models trained on annotated corpora that are heterogeneous and frequently include critical SDOH. This work presents a new corpus with SDOH annotations, a novel active learning framework, and the first extraction results on the new corpus. The Social History Annotation Corpus (SHAC) includes 4,480 social history sections with detailed annotation for 12 SDOH characterizing the status, extent, and temporal information of 18K distinct events. We introduce a novel active learning framework that selects samples for annotation using a surrogate text classification task as a proxy for a more complex event extraction task. The active learning framework successfully increases the frequency of health risk factors and improves automatic extraction of these events over undirected annotation. An event extraction model trained on SHAC achieves high extraction performance for substance use status (0.82-0.93 F1), employment status (0.81-0.86 F1), and living status type (0.81-0.93 F1) on data from three institutions. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2004.05438v2-abstract-full').style.display = 'none'; document.getElementById('2004.05438v2-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, 2020; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 11 April, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2020. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> Journal of Biomedical Informatics 113 (2021) 103631 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1904.04398">arXiv:1904.04398</a> <span> [<a href="https://arxiv.org/pdf/1904.04398">pdf</a>, <a href="https://arxiv.org/format/1904.04398">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"> Disfluencies and Human Speech Transcription Errors </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zayats%2C+V">Vicky Zayats</a>, <a href="/search/cs?searchtype=author&query=Tran%2C+T">Trang Tran</a>, <a href="/search/cs?searchtype=author&query=Wright%2C+R">Richard Wright</a>, <a href="/search/cs?searchtype=author&query=Mansfield%2C+C">Courtney Mansfield</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1904.04398v1-abstract-short" style="display: inline;"> This paper explores contexts associated with errors in transcrip-tion of spontaneous speech, shedding light on human perceptionof disfluencies and other conversational speech phenomena. Anew version of the Switchboard corpus is provided with disfluency annotations for careful speech transcripts, together with results showing the impact of transcription errors on evaluation of automatic disfluency… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1904.04398v1-abstract-full').style.display = 'inline'; document.getElementById('1904.04398v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1904.04398v1-abstract-full" style="display: none;"> This paper explores contexts associated with errors in transcrip-tion of spontaneous speech, shedding light on human perceptionof disfluencies and other conversational speech phenomena. Anew version of the Switchboard corpus is provided with disfluency annotations for careful speech transcripts, together with results showing the impact of transcription errors on evaluation of automatic disfluency detection. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1904.04398v1-abstract-full').style.display = 'none'; document.getElementById('1904.04398v1-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 April, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 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">Submitted to INTERSPEECH 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/1904.04388">arXiv:1904.04388</a> <span> [<a href="https://arxiv.org/pdf/1904.04388">pdf</a>, <a href="https://arxiv.org/format/1904.04388">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"> Giving Attention to the Unexpected: Using Prosody Innovations in Disfluency Detection </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zayats%2C+V">Vicky Zayats</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1904.04388v1-abstract-short" style="display: inline;"> Disfluencies in spontaneous speech are known to be associated with prosodic disruptions. However, most algorithms for disfluency detection use only word transcripts. Integrating prosodic cues has proved difficult because of the many sources of variability affecting the acoustic correlates. This paper introduces a new approach to extracting acoustic-prosodic cues using text-based distributional pre… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1904.04388v1-abstract-full').style.display = 'inline'; document.getElementById('1904.04388v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1904.04388v1-abstract-full" style="display: none;"> Disfluencies in spontaneous speech are known to be associated with prosodic disruptions. However, most algorithms for disfluency detection use only word transcripts. Integrating prosodic cues has proved difficult because of the many sources of variability affecting the acoustic correlates. This paper introduces a new approach to extracting acoustic-prosodic cues using text-based distributional prediction of acoustic cues to derive vector z-score features (innovations). We explore both early and late fusion techniques for integrating text and prosody, showing gains over a high-accuracy text-only model. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1904.04388v1-abstract-full').style.display = 'none'; document.getElementById('1904.04388v1-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 April, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 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">Accepted at NAACL-HLT 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/1904.03296">arXiv:1904.03296</a> <span> [<a href="https://arxiv.org/pdf/1904.03296">pdf</a>, <a href="https://arxiv.org/format/1904.03296">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"> A General Framework for Information Extraction using Dynamic Span Graphs </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Luan%2C+Y">Yi Luan</a>, <a href="/search/cs?searchtype=author&query=Wadden%2C+D">Dave Wadden</a>, <a href="/search/cs?searchtype=author&query=He%2C+L">Luheng He</a>, <a href="/search/cs?searchtype=author&query=Shah%2C+A">Amy Shah</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Hajishirzi%2C+H">Hannaneh Hajishirzi</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="1904.03296v1-abstract-short" style="display: inline;"> We introduce a general framework for several information extraction tasks that share span representations using dynamically constructed span graphs. The graphs are constructed by selecting the most confident entity spans and linking these nodes with confidence-weighted relation types and coreferences. The dynamic span graph allows coreference and relation type confidences to propagate through the… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1904.03296v1-abstract-full').style.display = 'inline'; document.getElementById('1904.03296v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1904.03296v1-abstract-full" style="display: none;"> We introduce a general framework for several information extraction tasks that share span representations using dynamically constructed span graphs. The graphs are constructed by selecting the most confident entity spans and linking these nodes with confidence-weighted relation types and coreferences. The dynamic span graph allows coreference and relation type confidences to propagate through the graph to iteratively refine the span representations. This is unlike previous multi-task frameworks for information extraction in which the only interaction between tasks is in the shared first-layer LSTM. Our framework significantly outperforms the state-of-the-art on multiple information extraction tasks across multiple datasets reflecting different domains. We further observe that the span enumeration approach is good at detecting nested span entities, with significant F1 score improvement on the ACE dataset. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1904.03296v1-abstract-full').style.display = 'none'; document.getElementById('1904.03296v1-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 April, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 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">NAACL 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/1811.07236">arXiv:1811.07236</a> <span> [<a href="https://arxiv.org/pdf/1811.07236">pdf</a>, <a href="https://arxiv.org/format/1811.07236">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"> Robust cross-domain disfluency detection with pattern match networks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zayats%2C+V">Vicky Zayats</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1811.07236v1-abstract-short" style="display: inline;"> In this paper we introduce a novel pattern match neural network architecture that uses neighbor similarity scores as features, eliminating the need for feature engineering in a disfluency detection task. We evaluate the approach in disfluency detection for four different speech genres, showing that the approach is as effective as hand-engineered pattern match features when used on in-domain data a… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1811.07236v1-abstract-full').style.display = 'inline'; document.getElementById('1811.07236v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1811.07236v1-abstract-full" style="display: none;"> In this paper we introduce a novel pattern match neural network architecture that uses neighbor similarity scores as features, eliminating the need for feature engineering in a disfluency detection task. We evaluate the approach in disfluency detection for four different speech genres, showing that the approach is as effective as hand-engineered pattern match features when used on in-domain data and achieves superior performance in cross-domain scenarios. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1811.07236v1-abstract-full').style.display = 'none'; document.getElementById('1811.07236v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 17 November, 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">This paper was submitted to EMNLP 2018 and was rejected. Our EMNLP submission is posted here to establish concurrency with "Disfluency Detection using Auto-Correlational Neural Networks" by P. Lou, P. Anderson, M. Johnson which was submitted to EMNLP at the same time</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1808.09602">arXiv:1808.09602</a> <span> [<a href="https://arxiv.org/pdf/1808.09602">pdf</a>, <a href="https://arxiv.org/format/1808.09602">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"> Multi-Task Identification of Entities, Relations, and Coreference for Scientific Knowledge Graph Construction </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Luan%2C+Y">Yi Luan</a>, <a href="/search/cs?searchtype=author&query=He%2C+L">Luheng He</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Hajishirzi%2C+H">Hannaneh Hajishirzi</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="1808.09602v1-abstract-short" style="display: inline;"> We introduce a multi-task setup of identifying and classifying entities, relations, and coreference clusters in scientific articles. We create SciERC, a dataset that includes annotations for all three tasks and develop a unified framework called Scientific Information Extractor (SciIE) for with shared span representations. The multi-task setup reduces cascading errors between tasks and leverages c… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1808.09602v1-abstract-full').style.display = 'inline'; document.getElementById('1808.09602v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1808.09602v1-abstract-full" style="display: none;"> We introduce a multi-task setup of identifying and classifying entities, relations, and coreference clusters in scientific articles. We create SciERC, a dataset that includes annotations for all three tasks and develop a unified framework called Scientific Information Extractor (SciIE) for with shared span representations. The multi-task setup reduces cascading errors between tasks and leverages cross-sentence relations through coreference links. Experiments show that our multi-task model outperforms previous models in scientific information extraction without using any domain-specific features. We further show that the framework supports construction of a scientific knowledge graph, which we use to analyze information in scientific literature. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1808.09602v1-abstract-full').style.display = 'none'; document.getElementById('1808.09602v1-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 August, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2018. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> EMNLP 2018 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1808.08643">arXiv:1808.08643</a> <span> [<a href="https://arxiv.org/pdf/1808.08643">pdf</a>, <a href="https://arxiv.org/format/1808.08643">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Retrieval">cs.IR</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"> Scientific Relation Extraction with Selectively Incorporated Concept Embeddings </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Luan%2C+Y">Yi Luan</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Hajishirzi%2C+H">Hannaneh Hajishirzi</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="1808.08643v1-abstract-short" style="display: inline;"> This paper describes our submission for the SemEval 2018 Task 7 shared task on semantic relation extraction and classification in scientific papers. We extend the end-to-end relation extraction model of (Miwa and Bansal) with enhancements such as a character-level encoding attention mechanism on selecting pretrained concept candidate embeddings. Our official submission ranked the second in relatio… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1808.08643v1-abstract-full').style.display = 'inline'; document.getElementById('1808.08643v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1808.08643v1-abstract-full" style="display: none;"> This paper describes our submission for the SemEval 2018 Task 7 shared task on semantic relation extraction and classification in scientific papers. We extend the end-to-end relation extraction model of (Miwa and Bansal) with enhancements such as a character-level encoding attention mechanism on selecting pretrained concept candidate embeddings. Our official submission ranked the second in relation classification task (Subtask 1.1 and Subtask 2 Senerio 2), and the first in the relation extraction task (Subtask 2 Scenario 1). <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1808.08643v1-abstract-full').style.display = 'none'; document.getElementById('1808.08643v1-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 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/1806.02786">arXiv:1806.02786</a> <span> [<a href="https://arxiv.org/pdf/1806.02786">pdf</a>, <a href="https://arxiv.org/format/1806.02786">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"> Domain Adversarial Training for Accented Speech Recognition </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Sun%2C+S">Sining Sun</a>, <a href="/search/cs?searchtype=author&query=Yeh%2C+C">Ching-Feng Yeh</a>, <a href="/search/cs?searchtype=author&query=Hwang%2C+M">Mei-Yuh Hwang</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+L">Lei Xie</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="1806.02786v1-abstract-short" style="display: inline;"> In this paper, we propose a domain adversarial training (DAT) algorithm to alleviate the accented speech recognition problem. In order to reduce the mismatch between labeled source domain data ("standard" accent) and unlabeled target domain data (with heavy accents), we augment the learning objective for a Kaldi TDNN network with a domain adversarial training (DAT) objective to encourage the model… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1806.02786v1-abstract-full').style.display = 'inline'; document.getElementById('1806.02786v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1806.02786v1-abstract-full" style="display: none;"> In this paper, we propose a domain adversarial training (DAT) algorithm to alleviate the accented speech recognition problem. In order to reduce the mismatch between labeled source domain data ("standard" accent) and unlabeled target domain data (with heavy accents), we augment the learning objective for a Kaldi TDNN network with a domain adversarial training (DAT) objective to encourage the model to learn accent-invariant features. In experiments with three Mandarin accents, we show that DAT yields up to 7.45% relative character error rate reduction when we do not have transcriptions of the accented speech, compared with the baseline trained on standard accent data only. We also find a benefit from DAT when used in combination with training from automatic transcriptions on the accented data. Furthermore, we find that DAT is superior to multi-task learning for accented speech recognition. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1806.02786v1-abstract-full').style.display = 'none'; document.getElementById('1806.02786v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 7 June, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2018. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1806.02782">arXiv:1806.02782</a> <span> [<a href="https://arxiv.org/pdf/1806.02782">pdf</a>, <a href="https://arxiv.org/format/1806.02782">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="Machine Learning">cs.LG</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="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> Training Augmentation with Adversarial Examples for Robust Speech Recognition </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Sun%2C+S">Sining Sun</a>, <a href="/search/cs?searchtype=author&query=Yeh%2C+C">Ching-Feng Yeh</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Hwang%2C+M">Mei-Yuh Hwang</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+L">Lei Xie</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="1806.02782v2-abstract-short" style="display: inline;"> This paper explores the use of adversarial examples in training speech recognition systems to increase robustness of deep neural network acoustic models. During training, the fast gradient sign method is used to generate adversarial examples augmenting the original training data. Different from conventional data augmentation based on data transformations, the examples are dynamically generated bas… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1806.02782v2-abstract-full').style.display = 'inline'; document.getElementById('1806.02782v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1806.02782v2-abstract-full" style="display: none;"> This paper explores the use of adversarial examples in training speech recognition systems to increase robustness of deep neural network acoustic models. During training, the fast gradient sign method is used to generate adversarial examples augmenting the original training data. Different from conventional data augmentation based on data transformations, the examples are dynamically generated based on current acoustic model parameters. We assess the impact of adversarial data augmentation in experiments on the Aurora-4 and CHiME-4 single-channel tasks, showing improved robustness against noise and channel variation. Further improvement is obtained when combining adversarial examples with teacher/student training, leading to a 23% relative word error rate reduction on Aurora-4. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1806.02782v2-abstract-full').style.display = 'none'; document.getElementById('1806.02782v2-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 June, 2018; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 7 June, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2018. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1804.10202">arXiv:1804.10202</a> <span> [<a href="https://arxiv.org/pdf/1804.10202">pdf</a>, <a href="https://arxiv.org/format/1804.10202">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Human-Computer Interaction">cs.HC</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> </div> </div> <p class="title is-5 mathjax"> Sounding Board: A User-Centric and Content-Driven Social Chatbot </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Fang%2C+H">Hao Fang</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+H">Hao Cheng</a>, <a href="/search/cs?searchtype=author&query=Sap%2C+M">Maarten Sap</a>, <a href="/search/cs?searchtype=author&query=Clark%2C+E">Elizabeth Clark</a>, <a href="/search/cs?searchtype=author&query=Holtzman%2C+A">Ari Holtzman</a>, <a href="/search/cs?searchtype=author&query=Choi%2C+Y">Yejin Choi</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A. Smith</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1804.10202v1-abstract-short" style="display: inline;"> We present Sounding Board, a social chatbot that won the 2017 Amazon Alexa Prize. The system architecture consists of several components including spoken language processing, dialogue management, language generation, and content management, with emphasis on user-centric and content-driven design. We also share insights gained from large-scale online logs based on 160,000 conversations with real-wo… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1804.10202v1-abstract-full').style.display = 'inline'; document.getElementById('1804.10202v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1804.10202v1-abstract-full" style="display: none;"> We present Sounding Board, a social chatbot that won the 2017 Amazon Alexa Prize. The system architecture consists of several components including spoken language processing, dialogue management, language generation, and content management, with emphasis on user-centric and content-driven design. We also share insights gained from large-scale online logs based on 160,000 conversations with real-world users. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1804.10202v1-abstract-full').style.display = 'none'; document.getElementById('1804.10202v1-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, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 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">5 pages, 3 figures, NAACL 2018</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1804.09661">arXiv:1804.09661</a> <span> [<a href="https://arxiv.org/pdf/1804.09661">pdf</a>, <a href="https://arxiv.org/format/1804.09661">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="Information Retrieval">cs.IR</span> </div> </div> <p class="title is-5 mathjax"> Personalized Language Model for Query Auto-Completion </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jaech%2C+A">Aaron Jaech</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1804.09661v1-abstract-short" style="display: inline;"> Query auto-completion is a search engine feature whereby the system suggests completed queries as the user types. Recently, the use of a recurrent neural network language model was suggested as a method of generating query completions. We show how an adaptable language model can be used to generate personalized completions and how the model can use online updating to make predictions for users not… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1804.09661v1-abstract-full').style.display = 'inline'; document.getElementById('1804.09661v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1804.09661v1-abstract-full" style="display: none;"> Query auto-completion is a search engine feature whereby the system suggests completed queries as the user types. Recently, the use of a recurrent neural network language model was suggested as a method of generating query completions. We show how an adaptable language model can be used to generate personalized completions and how the model can use online updating to make predictions for users not seen during training. The personalized predictions are significantly better than a baseline that uses no user information. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1804.09661v1-abstract-full').style.display = 'none'; document.getElementById('1804.09661v1-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 April, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 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">ACL 2018</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1804.05499">arXiv:1804.05499</a> <span> [<a href="https://arxiv.org/pdf/1804.05499">pdf</a>, <a href="https://arxiv.org/ps/1804.05499">ps</a>, <a href="https://arxiv.org/format/1804.05499">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"> Community Member Retrieval on Social Media using Textual Information </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jaech%2C+A">Aaron Jaech</a>, <a href="/search/cs?searchtype=author&query=Hathi%2C+S">Shobhit Hathi</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1804.05499v1-abstract-short" style="display: inline;"> This paper addresses the problem of community membership detection using only text features in a scenario where a small number of positive labeled examples defines the community. The solution introduces an unsupervised proxy task for learning user embeddings: user re-identification. Experiments with 16 different communities show that the resulting embeddings are more effective for community member… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1804.05499v1-abstract-full').style.display = 'inline'; document.getElementById('1804.05499v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1804.05499v1-abstract-full" style="display: none;"> This paper addresses the problem of community membership detection using only text features in a scenario where a small number of positive labeled examples defines the community. The solution introduces an unsupervised proxy task for learning user embeddings: user re-identification. Experiments with 16 different communities show that the resulting embeddings are more effective for community membership identification than common unsupervised representations. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1804.05499v1-abstract-full').style.display = 'none'; document.getElementById('1804.05499v1-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 April, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 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">NAACL 2018</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1804.01189">arXiv:1804.01189</a> <span> [<a href="https://arxiv.org/pdf/1804.01189">pdf</a>, <a href="https://arxiv.org/format/1804.01189">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Systems and Control">eess.SY</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="Optimization and Control">math.OC</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"> Real-Time Prediction of the Duration of Distribution System Outages </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jaech%2C+A">Aaron Jaech</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+B">Baosen Zhang</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Kirschen%2C+D+S">Daniel S. Kirschen</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="1804.01189v2-abstract-short" style="display: inline;"> This paper addresses the problem of predicting duration of unplanned power outages, using historical outage records to train a series of neural network predictors. The initial duration prediction is made based on environmental factors, and it is updated based on incoming field reports using natural language processing to automatically analyze the text. Experiments using 15 years of outage records… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1804.01189v2-abstract-full').style.display = 'inline'; document.getElementById('1804.01189v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1804.01189v2-abstract-full" style="display: none;"> This paper addresses the problem of predicting duration of unplanned power outages, using historical outage records to train a series of neural network predictors. The initial duration prediction is made based on environmental factors, and it is updated based on incoming field reports using natural language processing to automatically analyze the text. Experiments using 15 years of outage records show good initial results and improved performance leveraging text. Case studies show that the language processing identifies phrases that point to outage causes and repair steps. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1804.01189v2-abstract-full').style.display = 'none'; document.getElementById('1804.01189v2-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 July, 2018; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 3 April, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 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">Appears in IEEE Transactions on Power Systems</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1710.02603">arXiv:1710.02603</a> <span> [<a href="https://arxiv.org/pdf/1710.02603">pdf</a>, <a href="https://arxiv.org/format/1710.02603">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"> Low-Rank RNN Adaptation for Context-Aware Language Modeling </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jaech%2C+A">Aaron Jaech</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1710.02603v2-abstract-short" style="display: inline;"> A context-aware language model uses location, user and/or domain metadata (context) to adapt its predictions. In neural language models, context information is typically represented as an embedding and it is given to the RNN as an additional input, which has been shown to be useful in many applications. We introduce a more powerful mechanism for using context to adapt an RNN by letting the context… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1710.02603v2-abstract-full').style.display = 'inline'; document.getElementById('1710.02603v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1710.02603v2-abstract-full" style="display: none;"> A context-aware language model uses location, user and/or domain metadata (context) to adapt its predictions. In neural language models, context information is typically represented as an embedding and it is given to the RNN as an additional input, which has been shown to be useful in many applications. We introduce a more powerful mechanism for using context to adapt an RNN by letting the context vector control a low-rank transformation of the recurrent layer weight matrix. Experiments show that allowing a greater fraction of the model parameters to be adjusted has benefits in terms of perplexity and classification for several different types of context. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1710.02603v2-abstract-full').style.display = 'none'; document.getElementById('1710.02603v2-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 May, 2018; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 6 October, 2017; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 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">Accepted to TACL</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1708.06075">arXiv:1708.06075</a> <span> [<a href="https://arxiv.org/pdf/1708.06075">pdf</a>, <a href="https://arxiv.org/format/1708.06075">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"> Scientific Information Extraction with Semi-supervised Neural Tagging </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Luan%2C+Y">Yi Luan</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Hajishirzi%2C+H">Hannaneh Hajishirzi</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="1708.06075v1-abstract-short" style="display: inline;"> This paper addresses the problem of extracting keyphrases from scientific articles and categorizing them as corresponding to a task, process, or material. We cast the problem as sequence tagging and introduce semi-supervised methods to a neural tagging model, which builds on recent advances in named entity recognition. Since annotated training data is scarce in this domain, we introduce a graph-ba… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1708.06075v1-abstract-full').style.display = 'inline'; document.getElementById('1708.06075v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1708.06075v1-abstract-full" style="display: none;"> This paper addresses the problem of extracting keyphrases from scientific articles and categorizing them as corresponding to a task, process, or material. We cast the problem as sequence tagging and introduce semi-supervised methods to a neural tagging model, which builds on recent advances in named entity recognition. Since annotated training data is scarce in this domain, we introduce a graph-based semi-supervised algorithm together with a data selection scheme to leverage unannotated articles. Both inductive and transductive semi-supervised learning strategies outperform state-of-the-art information extraction performance on the 2017 SemEval Task 10 ScienceIE task. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1708.06075v1-abstract-full').style.display = 'none'; document.getElementById('1708.06075v1-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 August, 2017; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 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">accepted by EMNLP 2017</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1704.07287">arXiv:1704.07287</a> <span> [<a href="https://arxiv.org/pdf/1704.07287">pdf</a>, <a href="https://arxiv.org/format/1704.07287">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="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Sound">cs.SD</span> </div> </div> <p class="title is-5 mathjax"> Parsing Speech: A Neural Approach to Integrating Lexical and Acoustic-Prosodic Information </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Tran%2C+T">Trang Tran</a>, <a href="/search/cs?searchtype=author&query=Toshniwal%2C+S">Shubham Toshniwal</a>, <a href="/search/cs?searchtype=author&query=Bansal%2C+M">Mohit Bansal</a>, <a href="/search/cs?searchtype=author&query=Gimpel%2C+K">Kevin Gimpel</a>, <a href="/search/cs?searchtype=author&query=Livescu%2C+K">Karen Livescu</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1704.07287v2-abstract-short" style="display: inline;"> In conversational speech, the acoustic signal provides cues that help listeners disambiguate difficult parses. For automatically parsing spoken utterances, we introduce a model that integrates transcribed text and acoustic-prosodic features using a convolutional neural network over energy and pitch trajectories coupled with an attention-based recurrent neural network that accepts text and prosodic… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1704.07287v2-abstract-full').style.display = 'inline'; document.getElementById('1704.07287v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1704.07287v2-abstract-full" style="display: none;"> In conversational speech, the acoustic signal provides cues that help listeners disambiguate difficult parses. For automatically parsing spoken utterances, we introduce a model that integrates transcribed text and acoustic-prosodic features using a convolutional neural network over energy and pitch trajectories coupled with an attention-based recurrent neural network that accepts text and prosodic features. We find that different types of acoustic-prosodic features are individually helpful, and together give statistically significant improvements in parse and disfluency detection F1 scores over a strong text-only baseline. For this study with known sentence boundaries, error analyses show that the main benefit of acoustic-prosodic features is in sentences with disfluencies, attachment decisions are most improved, and transcription errors obscure gains from prosody. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1704.07287v2-abstract-full').style.display = 'none'; document.getElementById('1704.07287v2-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 April, 2018; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 24 April, 2017; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 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">Accepted in NAACL HLT 2018</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1704.06380">arXiv:1704.06380</a> <span> [<a href="https://arxiv.org/pdf/1704.06380">pdf</a>, <a href="https://arxiv.org/format/1704.06380">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"> Improving Context Aware Language Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jaech%2C+A">Aaron Jaech</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1704.06380v1-abstract-short" style="display: inline;"> Increased adaptability of RNN language models leads to improved predictions that benefit many applications. However, current methods do not take full advantage of the RNN structure. We show that the most widely-used approach to adaptation (concatenating the context with the word embedding at the input to the recurrent layer) is outperformed by a model that has some low-cost improvements: adaptatio… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1704.06380v1-abstract-full').style.display = 'inline'; document.getElementById('1704.06380v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1704.06380v1-abstract-full" style="display: none;"> Increased adaptability of RNN language models leads to improved predictions that benefit many applications. However, current methods do not take full advantage of the RNN structure. We show that the most widely-used approach to adaptation (concatenating the context with the word embedding at the input to the recurrent layer) is outperformed by a model that has some low-cost improvements: adaptation of both the hidden and output layers. and a feature hashing bias term to capture context idiosyncrasies. Experiments on language modeling and classification tasks using three different corpora demonstrate the advantages of the proposed techniques. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1704.06380v1-abstract-full').style.display = 'none'; document.getElementById('1704.06380v1-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 April, 2017; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2017. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1704.06217">arXiv:1704.06217</a> <span> [<a href="https://arxiv.org/pdf/1704.06217">pdf</a>, <a href="https://arxiv.org/format/1704.06217">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"> Reinforcement Learning with External Knowledge and Two-Stage Q-functions for Predicting Popular Reddit Threads </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=He%2C+J">Ji He</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=He%2C+X">Xiaodong He</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="1704.06217v1-abstract-short" style="display: inline;"> This paper addresses the problem of predicting popularity of comments in an online discussion forum using reinforcement learning, particularly addressing two challenges that arise from having natural language state and action spaces. First, the state representation, which characterizes the history of comments tracked in a discussion at a particular point, is augmented to incorporate the global con… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1704.06217v1-abstract-full').style.display = 'inline'; document.getElementById('1704.06217v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1704.06217v1-abstract-full" style="display: none;"> This paper addresses the problem of predicting popularity of comments in an online discussion forum using reinforcement learning, particularly addressing two challenges that arise from having natural language state and action spaces. First, the state representation, which characterizes the history of comments tracked in a discussion at a particular point, is augmented to incorporate the global context represented by discussions on world events available in an external knowledge source. Second, a two-stage Q-learning framework is introduced, making it feasible to search the combinatorial action space while also accounting for redundancy among sub-actions. We experiment with five Reddit communities, showing that the two methods improve over previous reported results on this task. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1704.06217v1-abstract-full').style.display = 'none'; document.getElementById('1704.06217v1-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 April, 2017; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2017. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1704.02080">arXiv:1704.02080</a> <span> [<a href="https://arxiv.org/pdf/1704.02080">pdf</a>, <a href="https://arxiv.org/format/1704.02080">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"> Conversation Modeling on Reddit using a Graph-Structured LSTM </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zayats%2C+V">Vicky Zayats</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1704.02080v1-abstract-short" style="display: inline;"> This paper presents a novel approach for modeling threaded discussions on social media using a graph-structured bidirectional LSTM which represents both hierarchical and temporal conversation structure. In experiments with a task of predicting popularity of comments in Reddit discussions, the proposed model outperforms a node-independent architecture for different sets of input features. Analyses… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1704.02080v1-abstract-full').style.display = 'inline'; document.getElementById('1704.02080v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1704.02080v1-abstract-full" style="display: none;"> This paper presents a novel approach for modeling threaded discussions on social media using a graph-structured bidirectional LSTM which represents both hierarchical and temporal conversation structure. In experiments with a task of predicting popularity of comments in Reddit discussions, the proposed model outperforms a node-independent architecture for different sets of input features. Analyses show a benefit to the model over the full course of the discussion, improving detection in both early and late stages. Further, the use of language cues with the bidirectional tree state updates helps with identifying controversial comments. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1704.02080v1-abstract-full').style.display = 'none'; document.getElementById('1704.02080v1-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 April, 2017; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 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">Submitted to TACL</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1609.04779">arXiv:1609.04779</a> <span> [<a href="https://arxiv.org/pdf/1609.04779">pdf</a>, <a href="https://arxiv.org/format/1609.04779">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"> Characterizing the Language of Online Communities and its Relation to Community Reception </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Tran%2C+T">Trang Tran</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1609.04779v1-abstract-short" style="display: inline;"> This work investigates style and topic aspects of language in online communities: looking at both utility as an identifier of the community and correlation with community reception of content. Style is characterized using a hybrid word and part-of-speech tag n-gram language model, while topic is represented using Latent Dirichlet Allocation. Experiments with several Reddit forums show that style i… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1609.04779v1-abstract-full').style.display = 'inline'; document.getElementById('1609.04779v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1609.04779v1-abstract-full" style="display: none;"> This work investigates style and topic aspects of language in online communities: looking at both utility as an identifier of the community and correlation with community reception of content. Style is characterized using a hybrid word and part-of-speech tag n-gram language model, while topic is represented using Latent Dirichlet Allocation. Experiments with several Reddit forums show that style is a better indicator of community identity than topic, even for communities organized around specific topics. Further, there is a positive correlation between the community reception to a contribution and the style similarity to that community, but not so for topic similarity. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1609.04779v1-abstract-full').style.display = 'none'; document.getElementById('1609.04779v1-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 September, 2016; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2016. </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">EMNLP 2016</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1608.04808">arXiv:1608.04808</a> <span> [<a href="https://arxiv.org/pdf/1608.04808">pdf</a>, <a href="https://arxiv.org/format/1608.04808">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</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"> Learning Latent Local Conversation Modes for Predicting Community Endorsement in Online Discussions </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Fang%2C+H">Hao Fang</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+H">Hao Cheng</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</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="1608.04808v2-abstract-short" style="display: inline;"> Many social media platforms offer a mechanism for readers to react to comments, both positively and negatively, which in aggregate can be thought of as community endorsement. This paper addresses the problem of predicting community endorsement in online discussions, leveraging both the participant response structure and the text of the comment. The different types of features are integrated in a n… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1608.04808v2-abstract-full').style.display = 'inline'; document.getElementById('1608.04808v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1608.04808v2-abstract-full" style="display: none;"> Many social media platforms offer a mechanism for readers to react to comments, both positively and negatively, which in aggregate can be thought of as community endorsement. This paper addresses the problem of predicting community endorsement in online discussions, leveraging both the participant response structure and the text of the comment. The different types of features are integrated in a neural network that uses a novel architecture to learn latent modes of discussion structure that perform as well as deep neural networks but are more interpretable. In addition, the latent modes can be used to weight text features thereby improving prediction accuracy. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1608.04808v2-abstract-full').style.display = 'none'; document.getElementById('1608.04808v2-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, 2016; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 16 August, 2016; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2016. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">10 pages, 7 figures</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> SocialNLP Workshop at Conf. Empirical Methods Natural Language Process. (EMNLP), 2016 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1608.03030">arXiv:1608.03030</a> <span> [<a href="https://arxiv.org/pdf/1608.03030">pdf</a>, <a href="https://arxiv.org/format/1608.03030">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"> Hierarchical Character-Word Models for Language Identification </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jaech%2C+A">Aaron Jaech</a>, <a href="/search/cs?searchtype=author&query=Mulcaire%2C+G">George Mulcaire</a>, <a href="/search/cs?searchtype=author&query=Hathi%2C+S">Shobhit Hathi</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=Smith%2C+N+A">Noah A. Smith</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="1608.03030v1-abstract-short" style="display: inline;"> Social media messages' brevity and unconventional spelling pose a challenge to language identification. We introduce a hierarchical model that learns character and contextualized word-level representations for language identification. Our method performs well against strong base- lines, and can also reveal code-switching. </span> <span class="abstract-full has-text-grey-dark mathjax" id="1608.03030v1-abstract-full" style="display: none;"> Social media messages' brevity and unconventional spelling pose a challenge to language identification. We introduce a hierarchical model that learns character and contextualized word-level representations for language identification. Our method performs well against strong base- lines, and can also reveal code-switching. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1608.03030v1-abstract-full').style.display = 'none'; document.getElementById('1608.03030v1-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 August, 2016; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2016. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1606.03667">arXiv:1606.03667</a> <span> [<a href="https://arxiv.org/pdf/1606.03667">pdf</a>, <a href="https://arxiv.org/format/1606.03667">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Deep Reinforcement Learning with a Combinatorial Action Space for Predicting Popular Reddit Threads </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=He%2C+J">Ji He</a>, <a href="/search/cs?searchtype=author&query=Ostendorf%2C+M">Mari Ostendorf</a>, <a href="/search/cs?searchtype=author&query=He%2C+X">Xiaodong He</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+J">Jianshu Chen</a>, <a href="/search/cs?searchtype=author&query=Gao%2C+J">Jianfeng Gao</a>, <a href="/search/cs?searchtype=author&query=Li%2C+L">Lihong Li</a>, <a href="/search/cs?searchtype=author&query=Deng%2C+L">Li Deng</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="1606.03667v4-abstract-short" style="display: inline;"> We introduce an online popularity prediction and tracking task as a benchmark task for reinforcement learning with a combinatorial, natural language action space. A specified number of discussion threads predicted to be popular are recommended, chosen from a fixed window of recent comments to track. Novel deep reinforcement learning architectures are studied for effective modeling of the value fun… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1606.03667v4-abstract-full').style.display = 'inline'; document.getElementById('1606.03667v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1606.03667v4-abstract-full" style="display: none;"> We introduce an online popularity prediction and tracking task as a benchmark task for reinforcement learning with a combinatorial, natural language action space. A specified number of discussion threads predicted to be popular are recommended, chosen from a fixed window of recent comments to track. Novel deep reinforcement learning architectures are studied for effective modeling of the value function associated with actions comprised of interdependent sub-actions. The proposed model, which represents dependence between sub-actions through a bi-directional LSTM, gives the best performance across different experimental configurations and domains, and it also generalizes well with varying numbers of recommendation requests. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1606.03667v4-abstract-full').style.display = 'none'; document.getElementById('1606.03667v4-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 September, 2016; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 12 June, 2016; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2016. </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 be published in EMNLP 2016, 11 pages</span> </p> </li> </ol> <nav class="pagination is-small is-centered breathe-horizontal" role="navigation" aria-label="pagination"> <a href="" class="pagination-previous is-invisible">Previous </a> <a href="/search/?searchtype=author&query=Ostendorf%2C+M&start=50" class="pagination-next" >Next </a> <ul class="pagination-list"> <li> <a href="/search/?searchtype=author&query=Ostendorf%2C+M&start=0" class="pagination-link is-current" aria-label="Goto page 1">1 </a> </li> <li> <a href="/search/?searchtype=author&query=Ostendorf%2C+M&start=50" class="pagination-link " aria-label="Page 2" aria-current="page">2 </a> </li> </ul> </nav> <div class="is-hidden-tablet">  <span class="help" style="display: inline-block;"><a href="https://github.com/arXiv/arxiv-search/releases">Search v0.5.6 released 2020-02-24</a>  </span> </div> </div> </main> <footer> <div class="columns is-desktop" role="navigation" aria-label="Secondary">  <div class="column"> <div class="columns"> <div class="column"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/about">About</a></li> <li><a href="https://info.arxiv.org/help">Help</a></li> </ul> </div> <div class="column"> <ul class="nav-spaced"> <li> <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><title>contact arXiv</title><desc>Click here to contact arXiv</desc><path d="M502.3 190.8c3.9-3.1 9.7-.2 9.7 4.7V400c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V195.6c0-5 5.7-7.8 9.7-4.7 22.4 17.4 52.1 39.5 154.1 113.6 21.1 15.4 56.7 47.8 92.2 47.6 35.7.3 72-32.8 92.3-47.6 102-74.1 131.6-96.3 154-113.7zM256 320c23.2.4 56.6-29.2 73.4-41.4 132.7-96.3 142.8-104.7 173.4-128.7 5.8-4.5 9.2-11.5 9.2-18.9v-19c0-26.5-21.5-48-48-48H48C21.5 64 0 85.5 0 112v19c0 7.4 3.4 14.3 9.2 18.9 30.6 23.9 40.7 32.4 173.4 128.7 16.8 12.2 50.2 41.8 73.4 41.4z"/></svg> <a href="https://info.arxiv.org/help/contact.html"> Contact</a> </li> <li> <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><title>subscribe to arXiv mailings</title><desc>Click here to subscribe</desc><path d="M476 3.2L12.5 270.6c-18.1 10.4-15.8 35.6 2.2 43.2L121 358.4l287.3-253.2c5.5-4.9 13.3 2.6 8.6 8.3L176 407v80.5c0 23.6 28.5 32.9 42.5 15.8L282 426l124.6 52.2c14.2 6 30.4-2.9 33-18.2l72-432C515 7.8 493.3-6.8 476 3.2z"/></svg> <a href="https://info.arxiv.org/help/subscribe"> Subscribe</a> </li> </ul> </div> </div> </div>   <div class="column"> <div class="columns"> <div class="column"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/help/license/index.html">Copyright</a></li> <li><a href="https://info.arxiv.org/help/policies/privacy_policy.html">Privacy Policy</a></li> </ul> </div> <div class="column sorry-app-links"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/help/web_accessibility.html">Web Accessibility Assistance</a></li> <li> <p class="help"> <a class="a11y-main-link" href="https://status.arxiv.org" target="_blank">arXiv Operational Status <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 256 512" class="icon filter-dark_grey" role="presentation"><path d="M224.3 273l-136 136c-9.4 9.4-24.6 9.4-33.9 0l-22.6-22.6c-9.4-9.4-9.4-24.6 0-33.9l96.4-96.4-96.4-96.4c-9.4-9.4-9.4-24.6 0-33.9L54.3 103c9.4-9.4 24.6-9.4 33.9 0l136 136c9.5 9.4 9.5 24.6.1 34z"/></svg></a><br> Get status notifications via <a class="is-link" href="https://subscribe.sorryapp.com/24846f03/email/new" target="_blank"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><path d="M502.3 190.8c3.9-3.1 9.7-.2 9.7 4.7V400c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V195.6c0-5 5.7-7.8 9.7-4.7 22.4 17.4 52.1 39.5 154.1 113.6 21.1 15.4 56.7 47.8 92.2 47.6 35.7.3 72-32.8 92.3-47.6 102-74.1 131.6-96.3 154-113.7zM256 320c23.2.4 56.6-29.2 73.4-41.4 132.7-96.3 142.8-104.7 173.4-128.7 5.8-4.5 9.2-11.5 9.2-18.9v-19c0-26.5-21.5-48-48-48H48C21.5 64 0 85.5 0 112v19c0 7.4 3.4 14.3 9.2 18.9 30.6 23.9 40.7 32.4 173.4 128.7 16.8 12.2 50.2 41.8 73.4 41.4z"/></svg>email</a> or <a class="is-link" href="https://subscribe.sorryapp.com/24846f03/slack/new" target="_blank"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 448 512" class="icon filter-black" role="presentation"><path d="M94.12 315.1c0 25.9-21.16 47.06-47.06 47.06S0 341 0 315.1c0-25.9 21.16-47.06 47.06-47.06h47.06v47.06zm23.72 0c0-25.9 21.16-47.06 47.06-47.06s47.06 21.16 47.06 47.06v117.84c0 25.9-21.16 47.06-47.06 47.06s-47.06-21.16-47.06-47.06V315.1zm47.06-188.98c-25.9 0-47.06-21.16-47.06-47.06S139 32 164.9 32s47.06 21.16 47.06 47.06v47.06H164.9zm0 23.72c25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06H47.06C21.16 243.96 0 222.8 0 196.9s21.16-47.06 47.06-47.06H164.9zm188.98 47.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06h-47.06V196.9zm-23.72 0c0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06V79.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06V196.9zM283.1 385.88c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06v-47.06h47.06zm0-23.72c-25.9 0-47.06-21.16-47.06-47.06 0-25.9 21.16-47.06 47.06-47.06h117.84c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06H283.1z"/></svg>slack</a> </p> </li> </ul> </div> </div> </div>  </div> </footer> <script src="https://static.arxiv.org/static/base/1.0.0a5/js/member_acknowledgement.js"></script> </body> </html>

CINXE.COM

Search | arXiv e-print repository