BEGIN:VCALENDAR VERSION:2.0 PRODID:-//128.220.36.25//NONSGML kigkonsult.se iCalcreator 2.26.9// CALSCALE:GREGORIAN METHOD:PUBLISH X-FROM-URL:https://www.clsp.jhu.edu X-WR-TIMEZONE:America/New_York BEGIN:VTIMEZONE TZID:America/New_York X-LIC-LOCATION:America/New_York BEGIN:STANDARD DTSTART:20231105T020000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RDATE:20241103T020000 TZNAME:EST END:STANDARD BEGIN:DAYLIGHT DTSTART:20240310T020000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RDATE:20250309T020000 TZNAME:EDT END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT UID:ai1ec-20716@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nOver the last few years\, deep neural models have tak en over the field of natural language processing (NLP)\, brandishing great improvements on many of its sequence-level tasks. But the end-to-end natu re of these models makes it hard to figure out whether the way they repres ent individual words aligns with how language builds itself from the botto m up\, or how lexical changes in register and domain can affect the untest ed aspects of such representations.\nIn this talk\, I will present NYTWIT\ , a dataset created to challenge large language models at the lexical leve l\, tasking them with identification of processes leading to the formation of novel English words\, as well as with segmentation and recovery of the specific subclass of novel blends. I will then present XRayEmb\, a method which alleviates the hardships of processing these novelties by fitting a character-level encoder to the existing models’ subword tokenizers\; and conclude with a discussion of the drawbacks of current tokenizers’ vocabul ary creation schemes.\nBiography\nYuval Pinter is a Senior Lecturer in the Department of Computer Science at Ben-Gurion University of the Negev\, fo cusing on natural language processing. Yuval got his PhD at the Georgia In stitute of Technology School of Interactive Computing as a Bloomberg Data Science PhD Fellow. Before that\, he worked as a Research Engineer at Yaho o Labs and as a Computational Linguist at Ginger Software\, and obtained a n MA in Linguistics and a BSc in CS and Mathematics\, both from Tel Aviv U niversity. Yuval blogs (in Hebrew) about language matters on Dagesh Kal. DTSTART;TZID=America/New_York:20210910T120000 DTEND;TZID=America/New_York:20210910T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD SEQUENCE:0 SUMMARY:Yuval Pinter (Ben-Gurion University – Virtual Visit) “Challenging a nd Adapting NLP Models to Lexical Phenomena” URL:https://www.clsp.jhu.edu/events/yuval-pinter/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n
\\nAbstr act
\nOver the last few years\, deep neural models have tak en over the field of natural language processing (NLP)\, brandishing great improvements on many of its sequence-level tasks. But the end-to-end natu re of these models makes it hard to figure out whether the way they repres ent individual words aligns with how language builds itself from the botto m up\, or how lexical changes in register and domain can affect the untest ed aspects of such representations.
\nIn this talk\, I will present NYTWIT\, a dataset created to challenge large language models at the lexic al level\, tasking them with identification of processes leading to the fo rmation of novel English words\, as well as with segmentation and recovery of the specific subclass of novel blends. I will then present XRayEmb\, a method which alleviates the hardships of processing these novelties by fi tting a character-level encoder to the existing models’ subword tokenizers \; and conclude with a discussion of the drawbacks of current tokenizers’ vocabulary creation schemes.
\nBiography
\nYuval Pinter
is a Senior Lecturer in the Department of Computer Science at Ben-Gurion
University of the Negev\, focusing on natural language processing. Yuval got his PhD at the Georgia Institute of Tec
hnology School of Interactive Computing as a Bloomberg Data Science PhD Fe
llow. Before that\, he worked as a Research Engineer at Yahoo Labs and as
a Computational Linguist at Ginger Software\, and obtained an MA in Lingui
stics and a BSc in CS and Mathematics\, both from Tel Aviv University.
Abstr act
\nText simplification aims to help audiences read and u nderstand a piece of text through lexical\, syntactic\, and discourse modi fications\, while remaining faithful to its central idea and meaning. Than ks to large-scale parallel corpora derived from Wikipedia and News\, much of modern-day text simplification research focuses on sentence simplificat ion\, transforming original\, more complex sentences into simplified versi ons. In this talk\, I present new frontiers that focus on discourse operat ions. First\, we consider the challenging task of simplifying highly techn ical language\, in our case\, medical texts. We introduce a new corpus of parallel texts in English comprising technical and lay summaries of all pu blished evidence pertaining to different clinical topics. We then propose a new metric to quantify stylistic differentiates between the two\, and mo dels for paragraph-level simplification. Second\, we present the first dat a-driven study of inserting elaborations and explanations during simplific ation\, and illustrate the richness and complexities of this phenomenon. p>\n
Biography
\nAbstr act
\nRaytheon BBN participated in the IARPA MATERIAL progr am\, whose objective is to enable rapid development of language-independen t methods for cross-lingual information retrieval (CLIR). The challenging CLIR task of retrieving documents written (or spoken) in one language so t hat they satisfy an information need expressed in a different language is exacerbated by unique challenges posed by the MATERIAL program: limited tr aining data for automatic speech recognition and machine translation\, sca nt lexical resources\, non-standardized orthography\, etc. Furthermore\, t he format of the queries and the “Query-Weighted Value” performance measur e are non-standard and not previously studied in the IR community. In this talk\, we will describe the Raytheon BBN CLIR system\, which was successf ul at addressing the above challenges and unique characteristics of the pr ogram.
\nBiography
\nDamianos Karakos has been at Raytheon BBN for the past nine years\, wh ere he is currently a Senior Principal Engineer\, Research. Before that\, he was research faculty at Johns Hopkins University. He has worked on seve ral Government projects (e.g.\, DARPA GALE\, DARPA RATS\, IARPA BABEL\, IA RPA MATERIAL\, IARPA BETTER) and on a variety of HLT-related topics (e.g.\ , speech recognition\, speech activity detection\, keyword search\, inform ation retrieval). He has published more than 60 peer-reviewed papers. His research interests lie at the intersection of human language technology an d machine learning\, with an emphasis on statistical methods. He obtained a PhD in Electrical Engineering from the University of Maryland\, College Park\, in 2002.
\n\n
Abstr act
\nWhile there is a vast amount of text written about ne arly any topic\, this is often difficult for someone unfamiliar with a spe cific field to understand. Automated text simplification aims to reduce th e complexity of a document\, making it more comprehensible to a broader au dience. Much of the research in this field has traditionally focused on si mplification sub-tasks\, such as lexical\, syntactic\, or sentence-level s implification. However\, current systems struggle to consistently produce high-quality simplifications. Phrase-based models tend to make too many po or transformations\; on the other hand\, recent neural models\, while prod ucing grammatical output\, often do not make all needed changes to the ori ginal text. In this thesis\, I discuss novel approaches for improving lexi cal and sentence-level simplification systems. Regarding sentence simplifi cation models\, after noting that encouraging diversity at inference time leads to significant improvements\, I take a closer look at the idea of di versity and perform an exhaustive comparison of diverse decoding technique s on other generation tasks. I also discuss the limitations in the framing of current simplification tasks\, which prevent these models from yet bei ng practically useful. Thus\, I also propose a retrieval-based reformulati on of the problem. Specifically\, starting with a document\, I identify co ncepts critical to understanding its content\, and then retrieve documents relevant for each concept\, re-ranking them based on the desired complexi ty level.
\nBiography
\nI ’m a research scientist at the HLTCOE at Johns Hopkins University. My prim ary research interests are in language generation\, diverse and constraine d decoding\, and information retrieval. During my PhD I focused mainly on the task of text simplification\, and now am working on formulating struct ured prediction problems as end-to-end generation tasks. I received my PhD in July 2021 from the University of Pennsylvania with Chris Callison-Burc h and Marianna Apidianaki.
\n\n X-TAGS;LANGUAGE=en-US:2021\,Kriz\,October END:VEVENT BEGIN:VEVENT UID:ai1ec-21023@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nSpeech data is notoriously difficult to work with due to a variety of codecs\, lengths of recordings\, and meta-data formats. W e present Lhotse\, a speech data representation library that draws upon le ssons learned from Kaldi speech recognition toolkit and brings its concept s into the modern deep learning ecosystem. Lhotse provides a common JSON d escription format with corresponding Python classes and data preparation r ecipes for over 30 popular speech corpora. Various datasets can be easily combined together and re-purposed for different tasks. The library handles multi-channel recordings\, long recordings\, local and cloud storage\, la zy and on-the-fly operations amongst other features. We introduce Cut and CutSet concepts\, which simplify common data wrangling tasks for audio and help incorporate acoustic context of speech utterances. Finally\, we show how Lhotse leverages PyTorch data API abstractions and adopts them to han dle speech data for deep learning.\nBiography\nPiotr Zelasko is an assista nt research scientist in the Center for Language and Speech Processing (CL SP) who specializes in automatic speech recognition (ASR) and spoken langu age understanding (SLU). His current research focuses on applying multilin gual and crosslingual speech recognition systems to categorize the phoneti c inventory of a previously unknown language and on improving defenses aga inst adversarial attacks on both speaker identification and automatic spee ch recognition systems. He is also addressing the question of how to struc ture a spontaneous conversation into high-level semantic units such as dia log acts or topics. Finally\, he is working on Lhotse + K2\, the next-gene ration speech processing research software ecosystem. Before joining Johns Hopkins\, Zelasko worked as a machine learning consultant for Avaya (2017 -2019)\, and as a machine learning engineer for Techmo (2015-2017). Zelask o received his PhD (2019) in electronics engineering\, as well as his mast er’s (2014) and undergraduate degrees (2013) in acoustic engineering from AGH University of Science and Technology in Kraków\, Poland. DTSTART;TZID=America/New_York:20211029T120000 DTEND;TZID=America/New_York:20211029T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore MD 21218 SEQUENCE:0 SUMMARY:Piotr Zelasko (CLSP at JHU) “Lhotse: a speech data representation l ibrary for the modern deep learning ecosystem” URL:https://www.clsp.jhu.edu/events/piotr-zelasko-clsp-at-jhu-lhotse-a-spee ch-data-representation-library-for-the-modern-deep-learning-ecosystem/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n
\\nAbstr act
\nSpeech data is notoriously difficult t o work with due to a variety of codecs\, lengths of recordings\, and meta- data formats. We present Lhotse\, a speech data representation library tha t draws upon lessons learned from Kaldi speech recognition toolkit and bri ngs its concepts into the modern deep learning ecosystem. Lhotse provides a common JSON description format with corresponding Python classes and dat a preparation recipes for over 30 popular speech corpora. Various datasets can be easily combined together and re-purposed for different tasks. The library handles multi-channel recordings\, long recordings\, local and clo ud storage\, lazy and on-the-fly operations amongst other features. We int roduce Cut and CutSet concepts\, which simplify common data wrangling task s for audio and help incorporate acoustic context of speech utterances. Fi nally\, we show how Lhotse leverages PyTorch data API abstractions and ado pts them to handle speech data for deep learning.
\nB iography
\nPiotr Zelasko is an assistant research scientist in the Center for Language and Speech Processing (CLSP) who specializes i n automatic speech recognition (ASR) and spoken language understanding (SL U). His current research focuses on applying multilingual and crosslingual speech recognition systems to categorize the phonetic inventory of a prev iously unknown language and on improving defenses against adversarial atta cks on both speaker identification and automatic speech recognition system s. He is also addressing the question of how to structure a spontaneous co nversation into high-level semantic units such as dialog acts or topics. F inally\, he is working on Lhotse + K2\, the next-generation speech process ing research software ecosystem. Before joining Johns Hopkins\, Zelasko wo rked as a machine learning consultant for Avaya (2017-2019)\, and as a mac hine learning engineer for Techmo (2015-2017). Zelasko received his PhD (2 019) in electronics engineering\, as well as his master’s (2014) and under graduate degrees (2013) in acoustic engineering from AGH University of Sci ence and Technology in Kraków\, Poland.
\n X-TAGS;LANGUAGE=en-US:2021\,October\,Zelasko END:VEVENT BEGIN:VEVENT UID:ai1ec-22374@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nIn recent years\, the field of Natural Language Proce ssing has seen a profusion of tasks\, datasets\, and systems that facilita te reasoning about real-world situations through language (e.g.\, RTE\, MN LI\, COMET). Such systems might\, for example\, be trained to consider a s ituation where “somebody dropped a glass on the floor\,” and conclude it i s likely that “the glass shattered” as a result. In this talk\, I will dis cuss three pieces of work that revisit assumptions made by or about these systems. In the first work\, I develop a Defeasible Inference task\, which enables a system to recognize when a prior assumption it has made may no longer be true in light of new evidence it receives. The second work I wil l discuss revisits partial-input baselines\, which have highlighted issues of spurious correlations in natural language reasoning datasets and led t o unfavorable assumptions about models’ reasoning abilities. In particular \, I will discuss experiments that show models may still learn to reason i n the presence of spurious dataset artifacts. Finally\, I will touch on wo rk analyzing harmful assumptions made by reasoning models in the form of s ocial stereotypes\, particularly in the case of free-form generative reaso ning models.\nBiography\nRachel Rudinger is an Assistant Professor in the Department of Computer Science at the University of Maryland\, College Par k. She holds joint appointments in the Department of Linguistics and the I nstitute for Advanced Computer Studies (UMIACS). In 2019\, Rachel complete d her Ph.D. in Computer Science at Johns Hopkins University in the Center for Language and Speech Processing. From 2019-2020\, she was a Young Inves tigator at the Allen Institute for AI in Seattle\, and a visiting research er at the University of Washington. Her research interests include computa tional semantics\, common-sense reasoning\, and issues of social bias and fairness in NLP. DTSTART;TZID=America/New_York:20220916T120000 DTEND;TZID=America/New_York:20220916T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Rachel Rudinger (University of Maryland\, College Park) “Not So Fas t!: Revisiting Assumptions in (and about) Natural Language Reasoning” URL:https://www.clsp.jhu.edu/events/rachel-rudinger-university-of-maryland- college-park-not-so-fast-revisiting-assumptions-in-and-about-natural-langu age-reasoning/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nIn recent years\, the field of Natural Language Proce ssing has seen a profusion of tasks\, datasets\, and systems that facilita te reasoning about real-world situations through language (e.g.\, RTE\, MN LI\, COMET). Such systems might\, for example\, be trained to consider a s ituation where “somebody dropped a glass on the floor\,” and conclude it i s likely that “the glass shattered” as a result. In this talk\, I will dis cuss three pieces of work that revisit assumptions made by or about these systems. In the first work\, I develop a Defeasible Inference task\, which enables a system to recognize when a prior assumption it has made may no longer be true in light of new evidence it receives. The second work I wil l discuss revisits partial-input baselines\, which have highlighted issues of spurious correlations in natural language reasoning datasets and led t o unfavorable assumptions about models’ reasoning abilities. In particular \, I will discuss experiments that show models may still learn to reason i n the presence of spurious dataset artifacts. Finally\, I will touch on wo rk analyzing harmful assumptions made by reasoning models in the form of s ocial stereotypes\, particularly in the case of free-form generative reaso ning models.
\nBiography
\nRachel Rudinger is an Assistant Professor in the Department of Computer Science at the Unive rsity of Maryland\, College Park. She holds joint appointments in the Depa rtment of Linguistics and the Institute for Advanced Computer Studies (UMI ACS). In 2019\, Rachel completed her Ph.D. in Computer Science at Johns Ho pkins University in the Center for Language and Speech Processing. From 20 19-2020\, she was a Young Investigator at the Allen Institute for AI in Se attle\, and a visiting researcher at the University of Washington. Her res earch interests include computational semantics\, common-sense reasoning\, and issues of social bias and fairness in NLP.
\n X-TAGS;LANGUAGE=en-US:2022\,Rudinger\,September END:VEVENT BEGIN:VEVENT UID:ai1ec-22375@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nI will present our work on data augmentation using st yle transfer as a way to improve domain adaptation in sequence labeling ta sks. The target domain is social media data\, and the task is named entity recognition (NER). The premise is that we can transform the labelled out of domain data into something that stylistically is more closely related t o the target data. Then we can train a model on a combination of the gener ated data and the smaller amount of in domain data to improve NER predicti on performance. I will show recent empirical results on these efforts.\nIf time allows\, I will also give an overview of other research projects I’m currently leading at RiTUAL (Research in Text Understanding and Analysis of Language) lab. The common thread among all these research problems is t he scarcity of labeled data.\nBiography\nThamar Solorio is a Professor of Computer Science at the University of Houston (UH). She holds graduate deg rees in Computer Science from the Instituto Nacional de Astrofísica\, Ópti ca y Electrónica\, in Puebla\, Mexico. Her research interests include info rmation extraction from social media data\, enabling technology for code-s witched data\, stylistic modeling of text\, and more recently multimodal a pproaches for online content understanding. She is the director and founde r of the RiTUAL Lab at UH. She is the recipient of an NSF CAREER award for her work on authorship attribution\, and recipient of the 2014 Emerging L eader ABIE Award in Honor of Denice Denton. She is currently serving a sec ond term as an elected board member of the North American Chapter of the A ssociation of Computational Linguistics and was PC co-chair for NAACL 2019 . She recently joined the team of Editors in Chief for the ACL Rolling Rev iew (ARR) system. Her research is currently funded by the NSF and by ADOBE . DTSTART;TZID=America/New_York:20220923T120000 DTEND;TZID=America/New_York:20220923T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Thamar Solorio (University of Houston) “Style Transfer for Data Aug mentation in Sequence Labeling Tasks” URL:https://www.clsp.jhu.edu/events/thamar-solorio-university-of-houston-st yle-transfer-for-data-augmentation-in-sequence-labeling-tasks/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nI will present our work on data a ugmentation using style transfer as a way to improve domain adaptation in sequence labeling tasks. The target domain is social media data\, and the task is named entity recognition (NER). The premise is that we can transfo rm the labelled out of domain data into something that stylistically is mo re closely related to the target data. Then we can train a model on a comb ination of the generated data and the smaller amount of in domain data to improve NER prediction performance. I will show recent empirical results o n these efforts.
\nIf time allows\, I will also give an overview of other research projects I’m currently leading at RiTUA L (Research in Text Understanding and Analysis of Language) lab. The commo n thread among all these research problems is the scarcity of labeled data .
\nBiography
\nThamar Solorio is a Professor of Computer Science at the Univer sity of Houston (UH). She holds graduate degrees in Computer Science from the Instituto Nacional de Astrofísica\, Óptica y Electrónica\, in Puebla\, Mexico. Her research interests include information extraction from social media data\, enabling technology for code-switched data\, stylistic model ing of text\, and more recently multimodal approaches for online content u nderstanding. She is the director and founder of the RiTUAL Lab at UH. She is the recipient of an NSF CAREER award for her work on authorship attrib ution\, and recipient of the 2014 Emerging Leader ABIE Award in Honor of D enice Denton. She is currently serving a second term as an elected board m ember of the North American Chapter of the Association of Computational Li nguistics and was PC co-chair for NAACL 2019. She recently joined the team of Editors in Chief for the ACL Rolling Review (ARR) system. Her research is currently funded by the NSF and by ADOBE.
\n X-TAGS;LANGUAGE=en-US:2022\,September\,Solorio END:VEVENT BEGIN:VEVENT UID:ai1ec-22380@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nThe availability of large multilingual pre-trained la nguage models has opened up exciting pathways for developing NLP technolog ies for languages with scarce resources. In this talk I will advocate for the need to go beyond the most common languages in multilingual evaluation \, and on the challenges of handling new\, unseen-during-training language s and varieties. I will also share some of my experiences with working wit h indigenous and other endangered language communities and activists.\nBio graphy\n\nAntonios Anastasopoulos is an Assistant Professor in Computer Sc ience at George Mason University. In 2019\, Antonis received his PhD in Co mputer Science from the University of Notre Dame and then worked as a post doctoral researcher at the Language Technologies Institute at Carnegie Mel lon University. His research interests revolve around computational lingui stics and natural language processing with a focus on low-resource setting s\, endangered languages\, and cross-lingual learning.\n\n\n DTSTART;TZID=America/New_York:20220930T120000 DTEND;TZID=America/New_York:20220930T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Antonios Anastasopoulos (George Mason University) “NLP Beyond the T op-100 Languages” URL:https://www.clsp.jhu.edu/events/antonis-anastasopoulos-george-mason-uni versity/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nThe availability of large multilingual pre-trained la nguage models has opened up exciting pathways for developing NLP technolog ies for languages with scarce resources. In this talk I will advocate for the need to go beyond the most common languages in multilingual evaluation \, and on the challenges of handling new\, unseen-during-training language s and varieties. I will also share some of my experiences with working wit h indigenous and other endangered language communities and activists.
\nBiography
\nAntonios Anastasopoulos is an Assistant Professor in Compu ter Science at George Mason University. In 2019\, Antonis received his PhD in Computer Science from the University of Notre Dame and then worked as a postdoctoral researcher at the Language Technologies Institute at Carneg ie Mellon University. His research interests revolve around computational linguistics and natural language processing with a focus on low-resource s ettings\, endangered languages\, and cross-lingual learning.
\n\n X-TAGS;LANGUAGE=en-US:2022\,Anastasopoulos\,September END:VEVENT BEGIN:VEVENT UID:ai1ec-22423@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20221007T120000 DTEND;TZID=America/New_York:20221007T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Ariya Rastrow (Amazon) URL:https://www.clsp.jhu.edu/events/ariya-rastrow-amazon-2/ X-COST-TYPE:free X-TAGS;LANGUAGE=en-US:2022\,October\,Rastrow END:VEVENT BEGIN:VEVENT UID:ai1ec-22394@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\n\nModel robustness and spurious correlations have rec eived increasing attention in the NLP community\, both in methods and eval uation. The term “spurious correlation” is overloaded though and can refer to any undesirable shortcuts learned by the model\, as judged by domain e xperts.\n\n\nWhen designing mitigation algorithms\, we often (implicitly) assume that a spurious feature is irrelevant for prediction. However\, man y features in NLP (e.g. word overlap and negation) are not spurious in the sense that the background is spurious for classifying objects in an image . In contrast\, they carry important information that’s needed to make pre dictions by humans. In this talk\, we argue that it is more productive to characterize features in terms of their necessity and sufficiency for pred iction. We then discuss the implications of this categorization in represe ntation\, learning\, and evaluation.\nBiography\nHe He is an Assistant Pro fessor in the Department of Computer Science and the Center for Data Scien ce at New York University. She obtained her PhD in Computer Science at the University of Maryland\, College Park. Before joining NYU\, she spent a y ear at AWS AI and was a post-doc at Stanford University before that. She i s interested in building robust and trustworthy NLP systems in human-cente red settings. Her recent research focus includes robust language understan ding\, collaborative text generation\, and understanding capabilities and issues of large language models. DTSTART;TZID=America/New_York:20221014T120000 DTEND;TZID=America/New_York:20221014T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:He He (New York University) “What We Talk about When We Talk about Spurious Correlations in NLP” URL:https://www.clsp.jhu.edu/events/he-he-new-york-university/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\n
Abstr act
\nModel robustness and spuri ous correlations have received increasing attention in the NLP community\, both in methods and evaluation. The term “spurious correlation” is overlo aded though and can refer to any undesirable shortcuts learned by the mode l\, as judged by domain experts.
\nWhen designing mitigation algorithms\, we often (implicitly) assume that a spurious feature is irrelevant for prediction. However\, many features in NLP (e.g. word overlap and negation) are not spurious in the sense that the background is spurious for classifying objects in an image. In contra st\, they carry important information that’s needed to make predictions by humans. In this talk\, we argue that it is more productive to characteriz e features in terms of their necessity and sufficiency for prediction. We then discuss the implications of this categorization in representation\, l earning\, and evaluation.
\nBiography
\nHe He is an Assistant Professor in the Department of Computer Science and the C enter for Data Science at New York University. She obtained her PhD in Com puter Science at the University of Maryland\, College Park. Before joining NYU\, she spent a year at AWS AI and was a post-doc at Stanford Universit y before that. She is interested in building robust and trustworthy NLP sy stems in human-centered settings. Her recent research focus includes robus t language understanding\, collaborative text generation\, and understandi ng capabilities and issues of large language models.
\nAbstr act
\nAbstr act
\nModern learning architectures for natural language processing have been very successful in incorporating a huge amount of texts into their parameters. However\, by and large\, such models store and use knowledge in distributed and decentralized ways. This proves unreliable and makes the models ill-suited for knowledge-intensive tasks that require reasoning over factual information in linguistic expre ssions. In this talk\, I will give a few examples of exploring alternativ e architectures to tackle those challenges. In particular\, we can improve the performance of such (language) models by representing\, storing and a ccessing knowledge in a dedicated memory component.
\nThis talk is based on several joint works with Yury Zemlyanskiy (Goo gle Research)\, Michiel de Jong (USC and Google Research)\, William Cohen (Google Research and CMU) and our other collaborators in Google Research.< /p>\n
Biography
\nFei is a research scientist at Google Research. Before that\, he was a Professor of Computer Science at U niversity of Southern California. His primary research interests are machi ne learning and its application to various AI problems: speech and languag e processing\, computer vision\, robotics and recently weather forecast an d climate modeling. He has a PhD (2007) from Computer and Information Sc ience from U. of Pennsylvania and B.Sc and M.Sc in Biomedical Engineering from Southeast University (Nanjing\, China).
\n X-TAGS;LANGUAGE=en-US:2022\,October\,Sha END:VEVENT BEGIN:VEVENT UID:ai1ec-23882@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nLarge language models (LLMs) have demonstrated incred ible power\, but they also possess vulnerabilities that can lead to misuse and potential attacks. In this presentation\, we will address two fundame ntal questions regarding the responsible utilization of LLMs: (1) How can we accurately identify AI-generated text? (2) What measures can safeguard the intellectual property of LLMs? We will introduce two recent watermarki ng techniques designed for text and models\, respectively. Our discussion will encompass the theoretical underpinnings that ensure the correctness o f watermark detection\, along with robustness against evasion attacks. Fur thermore\, we will showcase empirical evidence validating their effectiven ess. These findings establish a solid technical groundwork for policymaker s\, legal professionals\, and generative AI practitioners alike.\nBiograph y\nLei Li is an Assistant Professor in Language Technology Institute at Ca rnegie Mellon University. He received Ph.D. from Carnegie Mellon Universit y School of Computer Science. He is a recipient of ACL 2021 Best Paper Awa rd\, CCF Young Elite Award in 2019\, CCF distinguished speaker in 2017\, W u Wen-tsün AI prize in 2017\, and 2012 ACM SIGKDD dissertation award (runn er-up)\, and is recognized as Notable Area Chair of ICLR 2023. Previously\ , he was a faculty member at UC Santa Barbara. Prior to that\, he founded ByteDance AI Lab in 2016 and led its research in NLP\, ML\, Robotics\, an d Drug Discovery. He launched ByteDance’s machine translation system VolcT rans and AI writing system Xiaomingbot\, serving one billion users. DTSTART;TZID=America/New_York:20230901T120000 DTEND;TZID=America/New_York:20230901T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Lei Li (Carnegie Mellon University) “Empowering Responsible Use of Large Language Models” URL:https://www.clsp.jhu.edu/events/lei-li-carnegie-mellon-university-empow ering-responsible-use-of-large-language-models/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nLarge language models (LLMs) have demonstrated incred ible power\, but they also possess vulnerabilities that can lead to misuse and potential attacks. In this presentation\, we will address two fundame ntal questions regarding the responsible utilization of LLMs: (1) How can we accurately identify AI-generated text? (2) What measures can safeguard the intellectual property of LLMs? We will introduce two recent watermarki ng techniques designed for text and models\, respectively. Our discussion will encompass the theoretical underpinnings that ensure the correctness o f watermark detection\, along with robustness against evasion attacks. Fur thermore\, we will showcase empirical evidence validating their effectiven ess. These findings establish a solid technical groundwork for policymaker s\, legal professionals\, and generative AI practitioners alike.
\n< strong>Biography
\nLei Li is an Assistant Professor in Lang uage Technology Institute at Carnegie Mellon University. He received Ph.D. from Carnegie Mellon University School of Computer Science. He is a recip ient of ACL 2021 Best Paper Award\, CCF Young Elite Award in 2019\, CCF di stinguished speaker in 2017\, Wu Wen-tsün AI prize in 2017\, and 2012 ACM SIGKDD dissertation award (runner-up)\, and is recognized as Notable Area Chair of ICLR 2023. Previously\, he was a faculty member at UC Santa Barba ra. Prior to that\, he founded ByteDance AI Lab in 2016 and led its resea rch in NLP\, ML\, Robotics\, and Drug Discovery. He launched ByteDance’s m achine translation system VolcTrans and AI writing system Xiaomingbot\, se rving one billion users.
\n X-TAGS;LANGUAGE=en-US:2023\,Li\,September END:VEVENT BEGIN:VEVENT UID:ai1ec-23886@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nThe arms race to build increasingly larger\, powerful language models (LMs) in the past year has been remarkable. Yet incorpora ting LMs effectively into practical applications that facilitate manual wo rkflows remains challenging. I will discuss LMs’ limiting factors and our efforts to overcome them. I will start with challenges surrounding efficie nt and robust LM alignment. I will share insights from our recent paper “S elf-Instruct” (ACL 2023)\, where we used vanilla (unaligned) LMs for align ing itself\, an approach that has yielded some success. Then\, I will move on to the challenge of tracing the output of LMs to reliable sources\, a weakness that makes them prone to hallucinations. I will discuss our recen t approach of ‘according-to’ prompting\, which steers LMs to quote directl y from sources observed in its pre-training. If time permits\, I will disc uss our ongoing project to adapt LMs to interact with web pages. Throughou t the presentation\, I will highlight our progress\, and end with question s about our future progress.\nBiography\nDaniel Khashabi is an assistant p rofessor in computer science at Johns Hopkins University and the Center fo r Language and Speech Processing (CLSP) member. He is interested in buildi ng reasoning-driven modular NLP systems that are robust\, transparent\, an d communicative\, particularly those that use natural language as the comm unication medium. Khashabi has published over 40 papers on natural languag e processing and AI in top-tier venues. His work touches upon developing. His research has won the ACL 2023 Outstanding Paper Award\, NAACL 2022 Bes t Paper Award\, research gifts from the Allen Institute for AI\, and an Am azon Research Award 2023. Before joining Hopkins\, he was a postdoctoral f ellow at the Allen Institute for AI (2019-2022) and obtained a Ph.D. from the University of Pennsylvania in 2019. DTSTART;TZID=America/New_York:20230908T120000 DTEND;TZID=America/New_York:20230908T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Daniel Khashabi (Johns Hopkins University) “Building More Helpful L anguage Models” URL:https://www.clsp.jhu.edu/events/daniel-khashabi-johns-hopkins-universit y/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nThe arms race to build increasingly larger\, powerful language models (LMs) in the past year has been remarkable. Yet incorpora ting LMs effectively into practical applications that facilitate manual wo rkflows remains challenging. I will discuss LMs’ limiting factors and our efforts to overcome them. I will start with challenges surrounding efficie nt and robust LM alignment. I will share insights from our recent paper “Self-Instruct” (ACL 2023)\, where we used vanilla (unaligned) LMs for aligning itself\, an approach that has yielded some success. Then\, I will move on to the challenge of t racing the output of LMs to reliable sources\, a weakness that makes them prone to hallucinations. I will discuss our recent approach of ‘according-to’ prompting\, which steers LM s to quote directly from sources observed in its pre-training. If time per mits\, I will discuss our ongoing project to adapt LMs to interact with we b pages. Throughout the presentation\, I will highlight our progress\, and end with questions about our future progress.
\nBiography strong>
\nDaniel Khashabi is an assistant professor in computer science at Johns Hopkins University and the Center for Language and Speech Pr ocessing (CLSP) member. He is interested in building reasoning-driven modu lar NLP systems that are robust\, transparent\, and communicative\, partic ularly those that use natural language as the communication medium. Khasha bi has published over 40 papers on natural language processing and AI in t op-tier venues. His work touches upon developing. His research has won the ACL 2023 Outstanding Paper Award\, NAACL 2022 Best Paper Award\, research gifts from the Allen Institute for AI\, and an Amazon Research Award 2023 . Before joining Hopkins\, he was a postdoctoral fellow at the Allen Insti tute for AI (2019-2022) and obtained a Ph.D. from the University of Pennsy lvania in 2019.
\n X-TAGS;LANGUAGE=en-US:2023\,Khashabi\,September END:VEVENT BEGIN:VEVENT UID:ai1ec-23888@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Student Seminars CONTACT: DESCRIPTION:Abstract\nEmbedding text sequences is a widespread requirement in modern language understanding. Existing approaches focus largely on con stant-size representations. This is problematic\, as the amount of informa tion contained in text often varies with the length of the input. We propo se a solution called Nugget\, which encodes language into a representation based on a dynamically selected subset of input tokens. These nuggets are learned through tasks like autoencoding and machine translation\, and int uitively segment language into meaningful units. We demonstrate Nugget out performs related approaches in tasks involving semantic comparison. Finall y\, we illustrate these compact units allow for expanding the contextual w indow of a language model (LM)\, suggesting new future LMs that can condit ion on significantly larger amounts of content. DTSTART;TZID=America/New_York:20230911T120000 DTEND;TZID=America/New_York:20230911T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Student Seminar – Guanghui Qin “Nugget: Neural Agglomerative Embedd ings of Text (ICML 2023)” URL:https://www.clsp.jhu.edu/events/student-seminar-guanghui-qin/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nEmbedding text sequ ences is a widespread requirement in modern language understanding. Existi ng approaches focus largely on constant-size representations. This is prob lematic\, as the amount of information contained in text often varies with the length of the input. We propose a solution called Nugget\, which enco des language into a representation based on a dynamically selected subset of input tokens. These nuggets are learned through tasks like autoencoding and machine translation\, and intuitively segment language into meaningfu l units. We demonstrate Nugget outperforms related approaches in tasks inv olving semantic comparison. Finally\, we illustrate these compact units al low for expanding the contextual window of a language model (LM)\, suggest ing new future LMs that can condition on significantly larger amounts of c ontent.
\n X-TAGS;LANGUAGE=en-US:2023\,Qin\,September END:VEVENT BEGIN:VEVENT UID:ai1ec-23892@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nThe growing power in computing and AI promises a near -term future of human-machine teamwork. In this talk\, I will present my r esearch group’s efforts in understanding the complex dynamics of human-mac hine interaction and designing intelligent machines aimed to assist and co llaborate with people. I will focus on 1) tools for onboarding machine tea mmates and authoring machine assistance\, 2) methods for detecting\, and b roadly managing\, errors in collaboration\, and 3) building blocks of know ledge needed to enable ad hoc human-machine teamwork. I will also highligh t our recent work on designing assistive\, collaborative machines to suppo rt older adults aging in place.\nBiography\nChien-Ming Huang is the John C . Malone Assistant Professor in the Department of Computer Science at the Johns Hopkins University. His research focuses on designing interactive AI aimed to assist and collaborate with people. He publishes in top-tier ven ues in HRI\, HCI\, and robotics including Science Robotics\, HRI\, CHI\, a nd CSCW. His research has received media coverage from MIT Technology Revi ew\, Tech Insider\, and Science Nation. Huang completed his postdoctoral t raining at Yale University and received his Ph.D. in Computer Science at t he University of Wisconsin–Madison. He is a recipient of the NSF CAREER aw ard. https://www.cs.jhu.edu/~cmhuang/ DTSTART;TZID=America/New_York:20230915T120000 DTEND;TZID=America/New_York:20230915T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Chien-Ming Huang (Johns Hopkins University) “Becoming Teammates: De signing Assistive\, Collaborative Machines” URL:https://www.clsp.jhu.edu/events/chien-ming-huang-johns-hopkins-universi ty/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nThe growing power in computing and AI promises a near -term future of human-machine teamwork. In this talk\, I will present my r esearch group’s efforts in understanding the complex dynamics of human-mac hine interaction and designing intelligent machines aimed to assist and co llaborate with people. I will focus on 1) tools for onboarding machine tea mmates and authoring machine assistance\, 2) methods for detecting\, and b roadly managing\, errors in collaboration\, and 3) building blocks of know ledge needed to enable ad hoc human-machine teamwork. I will also highligh t our recent work on designing assistive\, collaborative machines to suppo rt older adults aging in place.
\nBiography
\nChien-Ming Huang is the John C. Malone Assistant Professor in the Departm ent of Computer Science at the Johns Hopkins University. His research focu ses on designing interactive AI aimed to assist and collaborate with peopl e. He publishes in top-tier venues in HRI\, HCI\, and robotics including S cience Robotics\, HRI\, CHI\, and CSCW. His research has received media co verage from MIT Technology Review\, Tech Insider\, and Science Nation. Hua ng completed his postdoctoral training at Yale University and received his Ph.D. in Computer Science at the University of Wisconsin–Madison. He is a recipient of the NSF CAREER award. https://www .cs.jhu.edu/~cmhuang/
\n X-TAGS;LANGUAGE=en-US:2023\,Huang\,September END:VEVENT BEGIN:VEVENT UID:ai1ec-23894@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nThe use of NLP in the realm of financial technology i s broad and complex\, with applications ranging from sentiment analysis an d named entity recognition to question answering. Large Language Models (L LMs) have been shown to be effective on a variety of tasks\; however\, no LLM specialized for the financial domain has been reported in the literatu re. In this work\, we present BloombergGPT\, a 50 billion parameter langua ge model that is trained on a wide range of financial data. We construct a 363 billion token dataset based on Bloomberg’s extensive data sources\, p erhaps the largest domain-specific dataset yet\, augmented with 345 billio n tokens from general-purpose datasets. We validate BloombergGPT on stand ard LLM benchmarks\, open financial benchmarks\, and a suite of internal b enchmarks that most accurately reflect our intended usage. Our mixed datas et training leads to a model that outperforms existing models on financial tasks by significant margins without sacrificing performance on general L LM benchmarks. Additionally\, we explain our modeling choices\, training p rocess\, and evaluation methodology.\nBiography\nMark Dredze is the John C Malone Professor of Computer Science at Johns Hopkins University and the Director of Research (Foundations of AI) for the JHU AI-X Foundry. He deve lops Artificial Intelligence Systems based on natural language processing and explores applications to public health and medicine.\nProf. Dredze is affiliated with the Malone Center for Engineering in Healthcare\, the Cent er for Language and Speech Processing\, among others. He holds a joint app ointment in the Biomedical Informatics & Data Science Section (BIDS)\, und er the Department of Medicine (DOM)\, Division of General Internal Medicin e (GIM) in the School of Medicine. He obtained his PhD from the University of Pennsylvania in 2009. DTSTART;TZID=America/New_York:20230918T120000 DTEND;TZID=America/New_York:20230918T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Mark Dredze (Johns Hopkins University) “BloombergGPT: A Large Langu age Model for Finance” URL:https://www.clsp.jhu.edu/events/mark-dredze-johns-hopkins-university/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nThe use of NLP in the realm of financial technology i s broad and complex\, with applications ranging from sentiment analysis an d named entity recognition to question answering. Large Language Models (L LMs) have been shown to be effective on a variety of tasks\; however\, no LLM specialized for the financial domain has been reported in the literatu re. In this work\, we present BloombergGPT\, a 50 billion parameter langua ge model that is trained on a wide range of financial data. We construct a 363 billion token dataset based on Bloomberg’s extensive data sources\, p erhaps the largest domain-specific dataset yet\, augmented with 345 billio n tokens from general-purpose datasets. We validate BloombergGPT on stand ard LLM benchmarks\, open financial benchmarks\, and a suite of internal b enchmarks that most accurately reflect our intended usage. Our mixed datas et training leads to a model that outperforms existing models on financial tasks by significant margins without sacrificing performance on general L LM benchmarks. Additionally\, we explain our modeling choices\, training p rocess\, and evaluation methodology.
\nBiography
\nMark Dredze is the John C Malone Professor of Computer Science at Jo hns Hopkins University and the Director of Research (Foundations of AI) fo r the JHU AI-X Foundry. He develops Artificial Intelligence Systems based on natural language processing and explores applications to public health and medicine.
\nProf. Dredze is affiliated with the Malone Center fo r Engineering in Healthcare\, the Center for Language and Speech Processin g\, among others. He holds a joint appointment in the Bio medical Informatics & Data Science Section (< span class='il'>BIDS)\, under the Department of Medicine (DOM)\, Di vision of General Internal Medicine (GIM) in the School of Medicine. He ob tained his PhD from the University of Pennsylvania in 2009.
\n HTML> X-TAGS;LANGUAGE=en-US:2023\,Dredze\,September END:VEVENT BEGIN:VEVENT UID:ai1ec-23983@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nVisually rich documents (scanned or digital) remain i mportant for many consumer and business use cases. During this talk we wil l share recent work from our team in the Document Intelligence Lab of Adob e Research to understand\, create\, and interact with these documents. Fi rst\, we’ll share a series of work on building models to decompose and und erstand the structure of documents to support use cases around document an alysis and accessibility. Next\, we’ll explore document semantic understan ding for a project where we convert natural language contract clauses to c ode to support business automation. Finally\, we’ll discuss DocEdit\, a mo del and dataset that enables editing structured documents from natural lan guage. \nBIOS:\nRajiv Jain is a Senior Research Scientist in the Document Intelligence Lab in Adobe Research\, where his research focuses on underst anding the layout\, content\, and interaction with documents. Prior to joi ning Adobe\, Rajiv was a consultant at DARPA\, where he worked on the Medi a Forensics Program to secure digital imagery. He previously served for 10 years as a researcher for the Department of Defense where he worked on pr ojects around large scale systems\, computer vision\, and network security . He received his PhD in computer science from the University of Maryland\ , College Park working in the field of document image analysis and retriev al.\nChris Tensmeyer primarily focuses on multi-modal document layout and content understanding as a Research Scientist in the Document Intelligence Lab of Adobe Research. Since joining Adobe 5 years ago\, his work has di rectly impacted popular Adobe features such as mobile Acrobat Liquid Mode\ , PDF table extraction\, handwriting recognition\, and scanned document de tection. Other research interests include general Computer Vision and Dee p Learning. He received his PhD in Computer Science from Brigham Young Un iversity on the topic of Deep Learning for Document Image Analysis. DTSTART;TZID=America/New_York:20230922T120000 DTEND;TZID=America/New_York:20230922T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Rajiv Jain and Chris Tensmeyer (Adobe) “Document Intelligence at Ad obe Research” URL:https://www.clsp.jhu.edu/events/rajiv-jain-and-chris-tensmeyer-adobe-do cument-intelligence-at-adobe-research/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nVisually rich document s (scanned or digital) remain important for many consumer and business use cases. During this talk we will sha re recent work from our team in the Document Intelligence Lab of Adobe Res earch to understand\, create\, and interact with these documents. First\, we’ll share a series of work on building models to decompose and understa nd the structure of documents to support use cases around document analysi s and accessibility. Next\, we’ll explore document semantic understanding for a project where we convert natural language contract clauses to code t o support business automation. Finally\, we’ll discuss DocEdit\, a model a nd dataset that enables editing structured documents from natural language .
\nBIOS:
\nRajiv Jain is a Senior Research Scientist in the Do cument Intelligence Lab in Adobe Research\, where his research focuses on understanding the layout\, content\, and interaction with documents. Prior to joining Adobe\, Rajiv was a consultant at DARPA\, where he worked on t he Media Forensics Program to secure digital imagery. He previously served for 10 years as a researcher for the Department of Defense where he worke d on projects around large scale systems\, computer vision\, and network s ecurity. He received his PhD in computer science from the University of Ma ryland\, College Park working in the field of document image analysis and retrieval.
\nChris Ten smeyer primarily focuses on multi-modal document layout and conte nt understanding as a Research Scientist in the Document Intelligence Lab of Adobe Research. Since joining Adobe 5 years ago\, his work has directl y impacted popular Adobe features such as mobile Acrobat Liquid Mode\, PDF table extraction\, handwriting recognition\, and scanned document detecti on. Other research interests include general Computer Vision and Deep Lea rning. He received his PhD in Computer Science from Brigham Young Univers ity on the topic of Deep Learning for Document Image Analysis.
\n X-TAGS;LANGUAGE=en-US:2023\,Jain and Tensmeyer\,September END:VEVENT BEGIN:VEVENT UID:ai1ec-23896@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nThe field of NLP is in the midst of a disruptive shif t\, fueled most recently by the advent of large language models (LLMs)\, w ith impacts on our methodologies\, funding and public perception. While th e core technologies and scope of real-world impact of our field may be cha nging (everything is different!)\, many of the same key challenges faced s ince the inception of our field remain (nothing has changed). In this talk I’ll describe recent work characterizing and tackling some of these chall enges\, notably: data-efficient domain adaptation and lifelong learning. I will also anchor discussion of cycles and shifts in the field by describi ng findings from a qualitative study of factors shaping the community over time\, including culture\, incentives\, and infrastructure. Through these complementary lenses into the past\, present and future\, I aim to inspir e shared hope\, excitement and discussion. \nBio\nEmma Strubell is the Raj Reddy Assistant Professor in the Language Technologies Institute in the S chool of Computer Science at Carnegie Mellon University\, and a Visiting S cientist at the Allen Institute for Artificial Intelligence. Previously sh e held research scientist roles at Google and FAIR after earning her docto ral degree in 2019 from the University of Massachusetts Amherst. Her resea rch lies at the intersection of natural language processing and machine le arning\, with a focus on providing pragmatic solutions to practitioners wh o wish to gain insights from natural language text via computation- and da ta-efficient AI. Her work has been recognized with a Madrona AI Impact Awa rd\, best paper awards at ACL and EMNLP\, and cited in news outlets includ ing the New York Times and Wall Street Journal. DTSTART;TZID=America/New_York:20230925T120000 DTEND;TZID=America/New_York:20230925T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Emma Strubell (Carnegie Mellon University) “Large Language Models: Everything’s Different and Nothing Has Changed” URL:https://www.clsp.jhu.edu/events/emma-strubell-carnegie-mellon-universit y/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nThe field of NLP i s in the midst of a disruptive shift\, fueled most recently by the advent of large language models (LLMs)\, with impacts on our methodologies\, fund ing and public perception. While the core technologies and scope of real-w orld impact of our field may be changing (everything is different!)\, many of the same key challenges faced since the inception of our field remain (nothing has changed). In this talk I’ll describe recent work characterizi ng and tackling some of these challenges\, notably: data-efficient domain adaptation and lifelong learning. I will also anchor discussion of cycles and shifts in the field by describing findings from a qualitative study of factors shaping the community over time\, including culture\, incentives\ , and infrastructure. Through these complementary lenses into the past\, p resent and future\, I aim to inspire shared hope\, excitement and discussi on.
\nBio
\n< span class='x_x_x_ContentPasted1'>Emma Strubell is the Raj Reddy Assistant Professor in the Language Technologies Institute in the School of Compute r Science at Carnegie Mellon University\, and a Visiting Scientist at the Allen Institute for Artificial Intelligence. Previously she held research scientist roles at Google and FAIR after earning her doctoral degree in 20 19 from the University of Massachusetts Amherst. Her research lies at the intersection of natural language processing and machine learning\, with a focus on providing pragmatic solutions to practitioners who wish to gain i nsights from natural language text via computation- and data-efficient AI. Her work has been recognized with a Madrona AI Impact Award\, best paper awards at ACL and EMNLP\, and cited in news outlets including the New York Times and Wall Street Journal.
\n X-TAGS;LANGUAGE=en-US:2023\,September\,Strubell END:VEVENT BEGIN:VEVENT UID:ai1ec-23898@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Student Seminars CONTACT: DESCRIPTION:Abstract\nAny valuable NLP dataset has traditionally been shipp ed with crowdsourced categorical labels. Instructions for collecting these labels are easy to communicate and the labels themselves are easy to anno tate. However\, as self-supervision based methods are getting better at ba sically everything\, human annotations may need to provide more nuanced su pervision or enable more detailed evaluation in order to be worth further collecting. One natural extension to existing categorical annotation schem es is to obtain uncertainty information beyond a single hard label. In thi s talk\, I will discuss my recent efforts on introducing scalar labels in place of categorical labels as a form of uncertainty annotation. We demons trate that\, compared to other more obvious annotation schemes for eliciti ng uncertainty information\, scalar labels are significantly more cost-eff ective to annotate\, provide reliable evaluation\, and have a theoretical connection to existing predictive uncertainty metrics. In particular\, the y motivate using other losses as surrogates for calibration evaluation. DTSTART;TZID=America/New_York:20230929T120000 DTEND;TZID=America/New_York:20230929T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:CLSP Student Seminar – Zhengping Jiang “Scalar Labels for Capturing Human Uncertainty” URL:https://www.clsp.jhu.edu/events/clsp-student-seminar-zhengping-jiang/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nAny valuable NLP d ataset has traditionally been shipped with crowdsourced categorical labels . Instructions for collecting these labels are easy to communicate and the labels themselves are easy to annotate. However\, as self-supervision bas ed methods are getting better at basically everything\, human annotations may need to provide more nuanced supervision or enable more detailed evalu ation in order to be worth further collecting. One natural extension to ex isting categorical annotation schemes is to obtain uncertainty information beyond a single hard label. In this talk\, I will discuss my recent effor ts on introducing scalar labels in place of categorical labels as a form o f uncertainty annotation. We demonstrate that\, compared to other more obv ious annotation schemes for eliciting uncertainty information\, scalar lab els are significantly more cost-effective to annotate\, provide reliable e valuation\, and have a theoretical connection to existing predictive uncer tainty metrics. In particular\, they motivate using other losses as surrog ates for calibration evaluation.
\n X-TAGS;LANGUAGE=en-US:2023\,Jiang\,September END:VEVENT BEGIN:VEVENT UID:ai1ec-23900@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Student Seminars CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20231002T120000 DTEND;TZID=America/New_York:20231002T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:CLSP Student Seminar – Anna Favaro URL:https://www.clsp.jhu.edu/events/clsp-student-seminar-anna-favaro/ X-COST-TYPE:free X-TAGS;LANGUAGE=en-US:2023\,Favaro\,October END:VEVENT BEGIN:VEVENT UID:ai1ec-24115@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Student Seminars CONTACT: DESCRIPTION:Abstract\nOur goal is to use AI to automatically find tax minim ization strategies\, an approach which we call “Shelter Check.” It would c ome in two variants. Existing-Authority Shelter Check would aim to find wh ether existing tax law authorities can be combined to create tax minimizat ion strategies\, so the IRS or Congress can shut them down. New-Authority Shelter Check would automate checking whether a new tax law authority – li ke proposed legislation or a draft court decision – would combine with exi sting authorities to create a new tax minimization strategy. We had initia lly had high hopes for GPT-* large language models for implementing Shelte r Check\, but our tests have showed that they do very poorly at basic lega l reasoning and handling legal text. So we are now creating a benchmark an d training data for LLM’s handling legal text\, hoping to spur improvement s. DTSTART;TZID=America/New_York:20231006T120000 DTEND;TZID=America/New_York:20231006T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:CLSP Student Seminar – Andrew Blair-Stanek “Shelter Check and GPT-4 ’s Bad Legal Performance” URL:https://www.clsp.jhu.edu/events/clsp-student-seminar-andrew-blair-stane k-shelter-check-and-gpt-4s-bad-legal-performance/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nOur goal is to use AI to automatically find tax minim ization strategies\, an approach which we call “Shelter Check.” It would c ome in two variants. Existing-Authority Shelter Check would aim to find wh ether existing tax law authorities can be combined to create tax minimizat ion strategies\, so the IRS or Congress can shut them down. New-Authority Shelter Check would automate checking whether a new tax law authority – li ke proposed legislation or a draft court decision – would combine with exi sting authorities to create a new tax minimization strategy. We had initia lly had high hopes for GPT-* large language models for implementing Shelte r Check\, but our tests have showed that they do very poorly at basic lega l reasoning and handling legal text. So we are now creating a benchmark an d training data for LLM’s handling legal text\, hoping to spur improvement s.
\n X-TAGS;LANGUAGE=en-US:2023\,Blair-Stanek\,October END:VEVENT BEGIN:VEVENT UID:ai1ec-24005@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nLarge-scale generative models such as GPT and DALL-E have revolutionized natural language processing and computer vision resear ch. These models not only generate high fidelity text or image outputs\, b ut also demonstrate impressive domain and task generalization capabilities . In contrast\, audio generative models are relatively primitive in scale and generalization.\nIn this talk\, I will start with a brief introduction on conventional neural speech generative models and discuss why they are unfit for scaling to Internet-scale data. Next\, by reviewing the latest l arge-scale generative models for text and image\, I will outline a few lin es of promising approaches to build scalable speech models. Last\, I will present Voicebox\, our latest work to advance this area. Voicebox is the m ost versatile generative model for speech. It is trained with a simple tas k — text conditioned speech infilling — on over 50K hours of multilingual speech with a powerful flow-matching objective. Through in-context learnin g\, Voicebox can perform monolingual/cross-lingual zero-shot TTS\, holisti c style conversion\, transient noise removal\, content editing\, and diver se sample generation. Moreover\, Voicebox achieves state-of-the-art perfor mance and excellent run-time efficiency.\nBiography\nWei-Ning Hsu is a res earch scientist at Meta Foundational AI Research (FAIR) and currently the lead of the audio generation team. His research focuses on self-supervised learning and generative models for speech and audio. His pioneering work includes HuBERT\, AV-HuBERT\, TextlessNLP\, data2vec\, wav2vec-U\, textles s speech translation\, and Voicebox. \nPrior to joining Meta\, Wei-Ning wo rked at MERL and Google Brain as a research intern. He received his Ph.D. and S.M. degrees in Electrical Engineering and Computer Science from Massa chusetts Institute of Technology in 2020 and 2018\, under the supervision of Dr. James Glass. He received his B.S. degree in Electrical Engineering from National Taiwan University in 2014\, under the supervision of Prof. L in-shan Lee and Prof. Hsuan-Tien Lin. DTSTART;TZID=America/New_York:20231009T120000 DTEND;TZID=America/New_York:20231009T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Wei-Ning Hsu (Meta Foundational AI Research) “Large Scale Universal Speech Generative Models” URL:https://www.clsp.jhu.edu/events/wei-ning-hsu-meta-foundational-ai-resea rch-large-scale-universal-speech-generative-models/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nLarge-scale generative models such as GPT and DALL-E have revolutionized natural langu age processing and computer vision research. These models not only generat e high fidelity text or image outputs\, but also demonstrate impressive do main and task generalization capabilities. In contrast\, audio generative models are relatively primitive in scale and generalization.
\nIn this talk\, I will st art with a brief introduction on conventional neural speech generative mod els and discuss why they are unfit for scaling to Internet-scale data. Nex t\, by reviewing the latest large-scale generative models for text and ima ge\, I will outline a few lines of promising approaches to build scalable speech models. Last\, I will present Voicebox\, our latest work to advance this area. Voicebox is the most versatile generative model for speech. It is trained with a simple task — text conditioned speech infilling — on ov er 50K hours of multilingual speech with a powerful flow-matching objectiv e. Through in-context learning\, Voicebox can perform monolingual/cross-li ngual zero-shot TTS\, holistic style conversion\, transient noise removal\ , content editing\, and diverse sample generation. Moreover\, Voicebox ach ieves state-of-the-art performance and excellent run-time efficiency.
\nBiography
\nWei-Ning Hsu is a research scientist at Meta Founda tional AI Research (FAIR) and currently the lead of the audio generation t eam. His research focuses on self-supervised learning and generative model s for speech and audio. His pioneering work includes HuBERT\, AV-HuBERT\, TextlessNLP\, data2vec\, wav2vec-U\, textless speech translation\, and Voi cebox.
\nPri or to joining Meta\, Wei-Ning worked at MERL and Google Brain as a researc h intern. He received his Ph.D. and S.M. degrees in Electrical Engineering and Computer Science from Massachusetts Institute of Technology in 2020 a nd 2018\, under the supervision of Dr. James Glass. He received his B.S. d egree in Electrical Engineering from National Taiwan University in 2014\, under the supervision of Prof. Lin-shan Lee and Prof. Hsuan-Tien Lin.
\n X-TAGS;LANGUAGE=en-US:2023\,Hsu\,October END:VEVENT BEGIN:VEVENT UID:ai1ec-23902@www.clsp.jhu.edu DTSTAMP:20240329T122330Z CATEGORIES;LANGUAGE=en-US:Seminars CONTACT: DESCRIPTION:Abstract\nPretrained language models (LMs) encode implicit repr esentations of knowledge in their parameters. Despite this observation\, o ur best methods for interpreting these representations yield few actionabl e insights on how to manipulate this parameter space for downstream benefi t. In this talk\, I will present work on methods that simulate machine rea soning by localizing and modifying parametric knowledge representations. F irst\, I will present a method for discovering knowledge-critical subnetwo rks within pretrained language models\, and show that these sparse computa tional subgraphs are responsible for the model’s ability to encode specifi c pieces of knowledge. Then\, I will present a new reasoning algorithm\, R ECKONING\, a bi-level optimisation procedure that dynamically encodes and reasons over new knowledge at test-time using the model’s existing learned knowledge representations as a scratchpad. Finally\, I will discuss next steps and challenges in using internal model mechanisms for reasoning.\n\n Bio\n\nAntoine Bosselut is an assistant professor in the School of Compute r and Communication Sciences at the École Polytechnique Fédéral de Lausann e (EPFL). He was a postdoctoral scholar at Stanford University and a Young Investigator at the Allen Institute for AI (AI2). He completed his PhD at the University of Washington and was a student researcher at Microsoft Re search. His research interests are in building systems that mix knowledge and language representations to solve problems in NLP\, specializing in co mmonsense representation and reasoning. DTSTART;TZID=America/New_York:20231013T120000 DTEND;TZID=America/New_York:20231013T131500 LOCATION:Hackerman Hall B17 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Antoine Bosselut (EPFL) “From Mechanistic Interpretability to Mecha nistic Reasoning” URL:https://www.clsp.jhu.edu/events/antoine-bosselut-epfl/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n\\nAbstr act
\nAbstr act
\nRecent advances in speech technology make heavy use o f pre-trained models that learn from large quantities of raw (untranscribe d) speech\, using “self-supervised” (ie unsupervised) learning. These mode ls learn to transform the acoustic input into a different representational format that makes supervised learning (for tasks such as transcription or even translation) much easier. However\, *what* and *how* speech-relevant information is encoded in these representations is not well understood. I will talk about some work at various stages of completion in which my gro up is analyzing the structure of these representations\, to gain a more sy stematic understanding of how word-level\, phonetic\, and speaker informat ion is encoded.
\nBiography
\nSharon Goldwate
r is a Professor in the Institute for Language\, Cognition and Computation
at the University of Edinburgh’s School of Informatics. She received her
PhD in 2007 from Brown University and spent two years as a postdoctoral re
searcher at Stanford University before moving to Edinburgh. Her research i
nterests include unsupervised and minimally-supervised learning for speech
and language processing\, computer modelling of language acquisition in c
hildren\, and computational studies of language use. Her main focus withi
n linguistics has been on the lower levels of structure including phonetic
s\, phonology\, and morphology.
Prof. Goldwater has received awards including the 2016 Roger Needha
m Award from the British Computer Society for “distinguished research cont
ribution in computer science by a UK-based researcher who has completed up
to 10 years of post-doctoral research.” She has served on the editorial b
oards of several journals\, including Computational Linguistics\, Transact
ions of the Association for Computational Linguistics\, and the inaugural
board of OPEN MIND: Advances in Cognitive Science. She was a program chair
for the EACL 2014 Conference and chaired the EACL governing board from 20
19-2020.