Profile picture

Vered Shwartz

Postdoctoral Researcher
Allen Institute for AI (AI2) and
Paul G. Allen School of Computer Science & Engineering
University of Washington
 v e r e d s at allenai dot org    @VeredShwartz     Blog     CV

My research interests focus on natural language processing, with the fundamental goal of building models capable of human-level understanding of natural language. I'm interested in computational semantics and pragmatics, and commonsense reasoning. I'm currently working on learning to uncover implicit meaning, which is abundant in human speech, and on developing machines with advanced reasoning skills.

I did my PhD (2019), M.Sc. (2015), and B.Sc. (2013) in Computer Science in Bar-Ilan University. I was part of the Natural Language Processing lab where I worked on lexical and compositional semantics.

☆ I'm looking for an academic position!

Research

robot cutting the branch it's sitting on

Commonsense Reasoning: We all share basic knowledge and reasoning ability regarding causes and effects, physical commonsense (if you cut the branch you sit on, you will fall), and social commonsense (it's impolite to comment on someone's weight). Endowing machine with such commonsense is challenging. Such knowledge is too vast to collect from humans and often too trivial to be mentioned in texts. In my work, I developed a model that actively seeks additional knowledge relevant for a given situation [1].

Check out our ACL 2020 commonsense reasoning tutorial!


Illustration of abductive reasoning by DELOREAN

Nonmonotonic Reasoning: Everyday causal reasoning requires reasoning about the plausible but potentially defeasible conclusions from incomplete or hypothetical observations. For example, abductive reasoning ("what might explain the current events?"), counterfactual reasoning ("what if?"), and defeasible reasoning ("what might weaken my conclusion?"). I'm working on developing systems capable of nonmonotonic reasoning for a wide range of situations describable in natural language [2, 3, 4].

 
baby scared of the interpretation of 'baby oil'

Lexical and Compositional Semantics: Lexical variability in human language, i.e. the ability to express the same meaning in various ways, is an obstacle for natural language understanding applications. Word representations excel at capturing topical similarity (elevator/floor), as well as functional similarity (elevator/escalator), but they lack the fine-grained distinction of the specific semantic relation between a pair of words. I developed methods for recognizing lexical semantic relations between words and phrases, including ontological relationships e.g., cat is a type of animal, tail is a part of cat [5, 6]; interpreting noun-compounds, e.g. olive oil is oil made of olives while baby oil is oil for babies [7, 8]; and identifying predicate paraphrases, e.g. that X die at Y may have the same meaning as X live until Y in certain contexts [9, 10].

 

Publications  

Filter by:

2021

Learning to Rationalize for Nonmonotonic Reasoning with Distant Supervision
Faeze Brahman, Vered Shwartz, Rachel Rudinger, and Yejin Choi. AAAI 2021.
preprint long conference

Paragraph-Level Commonsense Transformers with Recurrent Memory
Saadia Gabriel, Chandra Bhagavatula, Vered Shwartz, Ronan Le Bras, Maxwell Forbes, and Yejin Choi. AAAI 2021.
preprint long conference

2020

Do Neural Language Models Overcome Reporting Bias?
Vered Shwartz and Yejin Choi. COLING 2020.
paper bib code short conference poster

Unsupervised Commonsense Question Answering with Self-Talk
Vered Shwartz, Peter West, Ronan Le Bras, Chandra Bhagavatula, and Yejin Choi. EMNLP 2020.
paper bib code long conference slides demo video

"You are grounded!": Latent Name Artifacts in Pre-trained Language Models
Vered Shwartz, Rachel Rudinger, and Oyvind Tafjord. EMNLP 2020.
paper bib code short conference slides video

Back to the Future: Unsupervised Backprop-based Decoding for Counterfactual and Abductive Commonsense Reasoning
Lianhui Qin, Vered Shwartz, Peter West, Chandra Bhagavatula, Jena D. Hwang, Ronan Le Bras, Antoine Bosselut, and Yejin Choi.
EMNLP 2020.
paper bib long conference

Social Chemistry 101: Learning to Reason about Social and Moral Norms
Maxwell Forbes, Jena D. Hwang, Vered Shwartz, Maarten Sap, and Yejin Choi. EMNLP 2020.
paper bib long conference

Thinking Like a Skeptic: Defeasible Inference in Natural Language
Rachel Rudinger, Vered Shwartz, Jena D. Hwang, Chandra Bhagavatula, Maxwell Forbes,
Ronan Le Bras, Noah A. Smith, and Yejin Choi.
Findings of EMNLP 2020.
paper bib code long

Paraphrasing vs Coreferring: Two Sides of the Same Coin
Yehudit Meged, Avi Caciularu, Vered Shwartz, and Ido Dagan.
Findings of EMNLP 2020.
paper bib long

Introductory Tutorial: Commonsense Reasoning for Natural Language Processing
Maarten Sap, Vered Shwartz, Antoine Bosselut, Yejin Choi, and Dan Roth.
Tutorial @ ACL 2020.
paper

2019

Still a Pain in the Neck: Evaluating Text Representations on Lexical Composition
Vered Shwartz and Ido Dagan. TACL 2019.
paper bib code slides video long journal

Diversify Your Datasets: Analyzing Generalization via Controlled Variance in Adversarial Datasets
Ohad Rozen, Vered Shwartz, Roee Aharoni, and Ido Dagan. CoNLL 2019.
paper bib dataset long conference

Revisiting Joint Modeling of Cross-document Entity and Event Coreference Resolution
Shany Barhom, Vered Shwartz, Alon Eirew, Michael Bugert, Nils Reimers, and Ido Dagan. ACL 2019.
paper bib long conference

Evaluating Text GANs as Language Models
Guy Tevet, Gavriel Habib, Vered Shwartz, and Jonathan Berant. NAACL 2019.
paper bib short conference

PhD Dissertation: Learning High-Precision Lexical Inferences.
Vered Shwartz, BIU 2019.
pdf bib

A Systematic Comparison of English Noun Compound Representations
Vered Shwartz.
Proceedings of the Joint Workshop on Multiword Expressions and WordNet (MWE-WN 2019) @ ACL 2019.
paper bib code poster long

2018

Paraphrase to Explicate: Revealing Implicit Noun-Compound Relations
Vered Shwartz and Ido Dagan. ACL 2018.
paper bib code slides video long conference

Breaking NLI Systems with Sentences that Require Simple Lexical Inferences
Max Glockner, Vered Shwartz, and Yoav Goldberg. ACL 2018.
paper bib dataset slides video short conference

Olive Oil Is Made of Olives, Baby Oil Is Made for Babies:
Interpreting Noun Compounds Using Paraphrases in a Neural Model

Vered Shwartz and Chris Waterson. NAACL 2018.
paper bib code dataset poster short conference

Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment
Tu Vu and Vered Shwartz. *SEM 2018.
paper bib short conference

Neural Disambiguation of Causal Lexical Markers Based on Context
Eugenio Martinez Camara, Vered Shwartz, Iryna Gurevych, and Ido Dagan. IWCS 2017.
paper bib code long conference

SemEval-2018 Task 9: Hypernym Discovery
Jose Camacho-Collados, Claudio Delli Bovi, Luis Espinosa-Anke, Sergio Oramas, Tommaso Pasini, Enrico Santus,
Vered Shwartz, Roberto Navigli, and Horacio Saggion.
Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval) @ NAACL 2018.
paper bib long

2017

Acquiring Predicate Paraphrases from News Tweets
Vered Shwartz, Gabriel Stanovsky and Ido Dagan. *SEM 2017.
paper bib code poster short conference

Learning Antonyms with Paraphrases and a Morphology-aware Neural Network
Sneha Rajana, Chris Callison-Burch, Marianna Apidianaki and Vered Shwartz. *SEM 2017.
paper bib code long conference

Hypernyms under Siege: Linguistically-motivated Artillery for Hypernymy Detection
Vered Shwartz, Enrico Santus, and Dominik Schlechtweg. EACL 2017.
paper bib code slides long conference

A Consolidated Open Knowledge Representation for Multiple Texts
Rachel Wities, Vered Shwartz, Gabriel Stanovsky, Meni Adler, Ori Shapira, Shyam Upadhyay,
Dan Roth, Eugenio Martinez Camara, Iryna Gurevych and Ido Dagan.
Proceedings of the 2nd Workshop on Linking Models of Lexical,
Sentential and Discourse-level Semantics (LSDSem) @ EACL 2017.
paper bib code long

2016

Improving Hypernymy Detection with an Integrated Path-based and Distributional Method
Vered Shwartz, Yoav Goldberg and Ido Dagan. ACL 2016.
paper bib code dataset slides video long outstanding paper conference

Adding Context to Semantic Data-Driven Paraphrasing
Vered Shwartz and Ido Dagan. *SEM 2016.
paper bib dataset AMT templates & guidelines poster short best paper conference

Path-based vs. Distributional Information in Recognizing Lexical Semantic Relations
Vered Shwartz and Ido Dagan.
Proceedings of the 5th Workshop on Cognitive Aspects of the Lexicon (CogALex-V) @ COLING 2016.
paper bib code slides long

CogALex-V Shared Task:
LexNET - Integrated Path-based and Distributional Method for the Identification of Semantic Relations

Vered Shwartz and Ido Dagan.
Proceedings of the 5th Workshop on Cognitive Aspects of the Lexicon (CogALex-V) @ COLING 2016.
paper bib code slides long

2015

Learning to Exploit Structured Resources for Lexical Inference
Vered Shwartz, Omer Levy, Ido Dagan and Jacob Goldberger. CoNLL 2015.
paper bib code appendix dataset video poster long conference

Multi-Level Alignments As An Extensible Representation Basis for Textual Entailment Algorithms
Tae-Gil Noh, Sebastian Padó, Vered Shwartz, Ido Dagan, Vivi Nastase,
Kathrin Eichler, Lili Kotlerman and Meni Adler. *SEM 2015.
paper bib poster short conference
 

Talks  

2021

Recent Breakthroughs and Uphill Battles in Modern Natural Language Processing
January 2021, AI and Data Summit, Virtual
slides video (Hebrew)

2020

Unsupervised Methods for Commonsense Reasoning
October 2020, SRI Vision and Learning Seminar, Virtual
December 2020, University of Mannheim NLP Seminar, Virtual
slides

Recent Breakthroughs and Uphill Battles in Modern Natural Language Processing
October 2020, Global AI October Sessions, Virtual
August 2020, Ai4 Conference, Virtual
August 2020, Bayer Meets Academia, Virtual
slides video (Ai4)

What do I know? Pushing the boundaries of existing world knowledge
January 2020, Bar Ilan University, Ramat Gan
February 2020, Treehouse Meeting, Department of Linguistics, University of Washington, Seattle, Washington

2019

Not a piece of cake: on lexical composition and implicit information
December 2019, Workshop on Data-driven Approaches to Parsing and Semantic Composition, Tübingen, Germany
December 2019, Inria Lab, Paris, France
slides

Fast breaking and slow building of textual inference models
December 2019, Department of Linguistics, Universität Konstanz, Germany
slides

What do I know? Pushing the boundaries of existing world knowledge
October 2019, LTI Colloquium at Carnegie Mellon University, Pittsburgh, Pennsylvania

At Loose Ends: Challenges and Opportunities in Lexical Composition
January 2019, Machine Learning and Optimization Laboratory at EPFL, Lausanne, Switzerland
slides

How well can Neural Text Representations Address Lexical Composition?
January 2019, Applied Machine Learning Days (AMLD), Lausanne, Switzerland
slides

2018

Multi-word Units Under the Magnifying Glass
December 2018, ONLP Lab, Open University of Israel
slides

Are we there yet? Remaining Challenges in Deep Learning based Natural Language Processing
December 2018, WeAreDevelopers AI Congress, Vienna, Austria
slides video

Introduction to Natural Language Processing
June 2018, HALB Elementary School, Woodmere, New York
slides

Acquiring Lexical Semantic Knowledge
May 2018, Stanford University, Stanford, California
May 2018, University of Washington, Seattle, Washington
May 2018, Allen Institute for Artificial Intelligence (AI2), Seattle, Washington
June 2018, University of Pennsylvania, Philadelphia, Pennsylvania
June 2018, New York University, New York, New York
slides video (from AI2)

Paraphrase to Explicate: Revealing Implicit Noun-Compound Relations
May 2018, Google, Mountain View, California
slides

How are you two related? Corpus-based Learning of Lexical Semantic Relations
January 2018, Workshop on Wordnets and Word Embeddings, 9th Global Wordnet Conference, Singapore
slides

Acquiring Lexical Semantic Knowledge
January 2018, Computer Science Department Seminar, NUS School of Computing, Singapore
slides

2017

Acquiring Lexical Semantic Knowledge
December 2017, Advanced Topics NLP Seminar, Tel-Aviv University, Israel
slides

Acquiring Lexical Semantic Knowledge
November 2017, Google Research, Israel
slides

Recognizing Lexical Inference
August 2016, UKP lab, Darmstadt University, Germany
slides

2016

Recognizing Lexical Inference
April 2016, ProbModels Reading Group, EdinburghNLP, University of Edinburgh, Scotland
slides

Recognizing Lexical Inference
April 2016, Thomson Reuters, Israel
slides

Recognizing Lexical Inference
February 2016, Intel, Israel

2015

Learning To Exploit Structured Resources for Lexical Inference
April 2015, IBM Research, Israel

 

Resources  

Probably Approximately a Scientific Blog
Check out my blog where I explain NLP and machine learning in basic terms for non-experts.
blog

Introductory Tutorial on Commonsense Reasoning
ACL 2020, organized by: Maarten Sap, Vered Shwartz, Antoine Bosselut, Dan Roth, and Yejin Choi
website recording

Fundamentals of Deep Learning for Natural Language Processing workshop (via NVIDIA DLI)
February 2019 @ Bar-Ilan University
Introduction to NLP  Machine Translation  Text Classification  Word Embeddings

Presentation on Crowdsourcing in NLP
April 2018, Bar-Ilan University
slides