Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Noah A. Smith

Noah A. Smith

University of Washington, Allen Institute for Artificial Intelligence

On-the-Fly Controlled Text Generation with Experts and Anti-Experts


May 07, 2021
Alisa Liu, Maarten Sap, Ximing Lu, Swabha Swayamdipta, Chandra Bhagavatula, Noah A. Smith, Yejin Choi

Figure 1 for On-the-Fly Controlled Text Generation with Experts and Anti-Experts
Figure 2 for On-the-Fly Controlled Text Generation with Experts and Anti-Experts
Figure 3 for On-the-Fly Controlled Text Generation with Experts and Anti-Experts
Figure 4 for On-the-Fly Controlled Text Generation with Experts and Anti-Experts
* Accepted to ACL 2021, camera-ready version coming soon 
  Access Paper or Ask Questions

A Dataset of Information-Seeking Questions and Answers Anchored in Research Papers


May 07, 2021
Pradeep Dasigi, Kyle Lo, Iz Beltagy, Arman Cohan, Noah A. Smith, Matt Gardner

Figure 1 for A Dataset of Information-Seeking Questions and Answers Anchored in Research Papers
Figure 2 for A Dataset of Information-Seeking Questions and Answers Anchored in Research Papers
Figure 3 for A Dataset of Information-Seeking Questions and Answers Anchored in Research Papers
Figure 4 for A Dataset of Information-Seeking Questions and Answers Anchored in Research Papers
* Accepted at NAACL 2021; Project page: https://allenai.org/project/qasper 
  Access Paper or Ask Questions

Provable Limitations of Acquiring Meaning from Ungrounded Form: What will Future Language Models Understand?


Apr 22, 2021
William Merrill, Yoav Goldberg, Roy Schwartz, Noah A. Smith

Figure 1 for Provable Limitations of Acquiring Meaning from Ungrounded Form: What will Future Language Models Understand?
Figure 2 for Provable Limitations of Acquiring Meaning from Ungrounded Form: What will Future Language Models Understand?
Figure 3 for Provable Limitations of Acquiring Meaning from Ungrounded Form: What will Future Language Models Understand?
Figure 4 for Provable Limitations of Acquiring Meaning from Ungrounded Form: What will Future Language Models Understand?
* Accepted at TACL; pre-MIT Press publication version 
  Access Paper or Ask Questions

Go Forth and Prosper: Language Modeling with Ancient Textual History


Apr 18, 2021
Rik Koncel-Kedziorski, Noah A. Smith

Figure 1 for Go Forth and Prosper: Language Modeling with Ancient Textual History
Figure 2 for Go Forth and Prosper: Language Modeling with Ancient Textual History

  Access Paper or Ask Questions

Probing Across Time: What Does RoBERTa Know and When?


Apr 16, 2021
Leo Z. Liu, Yizhong Wang, Jungo Kasai, Hannaneh Hajishirzi, Noah A. Smith

Figure 1 for Probing Across Time: What Does RoBERTa Know and When?
Figure 2 for Probing Across Time: What Does RoBERTa Know and When?
Figure 3 for Probing Across Time: What Does RoBERTa Know and When?
Figure 4 for Probing Across Time: What Does RoBERTa Know and When?

  Access Paper or Ask Questions

Finetuning Pretrained Transformers into RNNs


Mar 24, 2021
Jungo Kasai, Hao Peng, Yizhe Zhang, Dani Yogatama, Gabriel Ilharco, Nikolaos Pappas, Yi Mao, Weizhu Chen, Noah A. Smith

Figure 1 for Finetuning Pretrained Transformers into RNNs
Figure 2 for Finetuning Pretrained Transformers into RNNs
Figure 3 for Finetuning Pretrained Transformers into RNNs
Figure 4 for Finetuning Pretrained Transformers into RNNs

  Access Paper or Ask Questions

Random Feature Attention


Mar 19, 2021
Hao Peng, Nikolaos Pappas, Dani Yogatama, Roy Schwartz, Noah A. Smith, Lingpeng Kong

Figure 1 for Random Feature Attention
Figure 2 for Random Feature Attention
Figure 3 for Random Feature Attention
Figure 4 for Random Feature Attention
* ICLR 2021 
  Access Paper or Ask Questions

Challenges in Automated Debiasing for Toxic Language Detection


Jan 29, 2021
Xuhui Zhou, Maarten Sap, Swabha Swayamdipta, Noah A. Smith, Yejin Choi

Figure 1 for Challenges in Automated Debiasing for Toxic Language Detection
Figure 2 for Challenges in Automated Debiasing for Toxic Language Detection
Figure 3 for Challenges in Automated Debiasing for Toxic Language Detection
Figure 4 for Challenges in Automated Debiasing for Toxic Language Detection
* EACL 2021 
  Access Paper or Ask Questions

GENIE: A Leaderboard for Human-in-the-Loop Evaluation of Text Generation


Jan 17, 2021
Daniel Khashabi, Gabriel Stanovsky, Jonathan Bragg, Nicholas Lourie, Jungo Kasai, Yejin Choi, Noah A. Smith, Daniel S. Weld

Figure 1 for GENIE: A Leaderboard for Human-in-the-Loop Evaluation of Text Generation
Figure 2 for GENIE: A Leaderboard for Human-in-the-Loop Evaluation of Text Generation
Figure 3 for GENIE: A Leaderboard for Human-in-the-Loop Evaluation of Text Generation
Figure 4 for GENIE: A Leaderboard for Human-in-the-Loop Evaluation of Text Generation

  Access Paper or Ask Questions

Shortformer: Better Language Modeling using Shorter Inputs


Dec 31, 2020
Ofir Press, Noah A. Smith, Mike Lewis

Figure 1 for Shortformer: Better Language Modeling using Shorter Inputs
Figure 2 for Shortformer: Better Language Modeling using Shorter Inputs
Figure 3 for Shortformer: Better Language Modeling using Shorter Inputs
Figure 4 for Shortformer: Better Language Modeling using Shorter Inputs

  Access Paper or Ask Questions

Infusing Finetuning with Semantic Dependencies


Dec 20, 2020
Zhaofeng Wu, Hao Peng, Noah A. Smith

Figure 1 for Infusing Finetuning with Semantic Dependencies
Figure 2 for Infusing Finetuning with Semantic Dependencies
Figure 3 for Infusing Finetuning with Semantic Dependencies
Figure 4 for Infusing Finetuning with Semantic Dependencies
* Accepted for publication in Transactions of the Association for Computational Linguistics (TACL), 2020. Author's final version (pre-MIT Press publication) 
  Access Paper or Ask Questions

Measuring Association Between Labels and Free-Text Rationales


Oct 24, 2020
Sarah Wiegreffe, Ana Marasovic, Noah A. Smith

Figure 1 for Measuring Association Between Labels and Free-Text Rationales
Figure 2 for Measuring Association Between Labels and Free-Text Rationales
Figure 3 for Measuring Association Between Labels and Free-Text Rationales
Figure 4 for Measuring Association Between Labels and Free-Text Rationales

  Access Paper or Ask Questions

Unsupervised Bitext Mining and Translation via Self-trained Contextual Embeddings


Oct 15, 2020
Phillip Keung, Julian Salazar, Yichao Lu, Noah A. Smith

Figure 1 for Unsupervised Bitext Mining and Translation via Self-trained Contextual Embeddings
Figure 2 for Unsupervised Bitext Mining and Translation via Self-trained Contextual Embeddings
Figure 3 for Unsupervised Bitext Mining and Translation via Self-trained Contextual Embeddings
Figure 4 for Unsupervised Bitext Mining and Translation via Self-trained Contextual Embeddings
* To appear in the Transactions of the Association for Computational Linguistics 
  Access Paper or Ask Questions

Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics


Oct 15, 2020
Swabha Swayamdipta, Roy Schwartz, Nicholas Lourie, Yizhong Wang, Hannaneh Hajishirzi, Noah A. Smith, Yejin Choi

Figure 1 for Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics
Figure 2 for Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics
Figure 3 for Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics
Figure 4 for Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics
* Proceedings of EMNLP 2020 
  Access Paper or Ask Questions

Natural Language Rationales with Full-Stack Visual Reasoning: From Pixels to Semantic Frames to Commonsense Graphs


Oct 15, 2020
Ana Marasović, Chandra Bhagavatula, Jae Sung Park, Ronan Le Bras, Noah A. Smith, Yejin Choi

Figure 1 for Natural Language Rationales with Full-Stack Visual Reasoning: From Pixels to Semantic Frames to Commonsense Graphs
Figure 2 for Natural Language Rationales with Full-Stack Visual Reasoning: From Pixels to Semantic Frames to Commonsense Graphs
Figure 3 for Natural Language Rationales with Full-Stack Visual Reasoning: From Pixels to Semantic Frames to Commonsense Graphs
Figure 4 for Natural Language Rationales with Full-Stack Visual Reasoning: From Pixels to Semantic Frames to Commonsense Graphs
* Accepted to Findings of EMNLP 
  Access Paper or Ask Questions

Plug and Play Autoencoders for Conditional Text Generation


Oct 12, 2020
Florian Mai, Nikolaos Pappas, Ivan Montero, Noah A. Smith, James Henderson

Figure 1 for Plug and Play Autoencoders for Conditional Text Generation
Figure 2 for Plug and Play Autoencoders for Conditional Text Generation
Figure 3 for Plug and Play Autoencoders for Conditional Text Generation
Figure 4 for Plug and Play Autoencoders for Conditional Text Generation
* To be published in EMNLP 2020 
  Access Paper or Ask Questions

The Multilingual Amazon Reviews Corpus


Oct 06, 2020
Phillip Keung, Yichao Lu, György Szarvas, Noah A. Smith

Figure 1 for The Multilingual Amazon Reviews Corpus
Figure 2 for The Multilingual Amazon Reviews Corpus
Figure 3 for The Multilingual Amazon Reviews Corpus
Figure 4 for The Multilingual Amazon Reviews Corpus
* To appear in EMNLP 2020 
  Access Paper or Ask Questions

Grounded Compositional Outputs for Adaptive Language Modeling


Oct 05, 2020
Nikolaos Pappas, Phoebe Mulcaire, Noah A. Smith

Figure 1 for Grounded Compositional Outputs for Adaptive Language Modeling
Figure 2 for Grounded Compositional Outputs for Adaptive Language Modeling
Figure 3 for Grounded Compositional Outputs for Adaptive Language Modeling
Figure 4 for Grounded Compositional Outputs for Adaptive Language Modeling
* EMNLP 2020 
  Access Paper or Ask Questions

Multilevel Text Alignment with Cross-Document Attention


Oct 03, 2020
Xuhui Zhou, Nikolaos Pappas, Noah A. Smith

Figure 1 for Multilevel Text Alignment with Cross-Document Attention
Figure 2 for Multilevel Text Alignment with Cross-Document Attention
Figure 3 for Multilevel Text Alignment with Cross-Document Attention
Figure 4 for Multilevel Text Alignment with Cross-Document Attention
* EMNLP 2020 
  Access Paper or Ask Questions

Parsing with Multilingual BERT, a Small Corpus, and a Small Treebank


Sep 29, 2020
Ethan C. Chau, Lucy H. Lin, Noah A. Smith

Figure 1 for Parsing with Multilingual BERT, a Small Corpus, and a Small Treebank
Figure 2 for Parsing with Multilingual BERT, a Small Corpus, and a Small Treebank
Figure 3 for Parsing with Multilingual BERT, a Small Corpus, and a Small Treebank
Figure 4 for Parsing with Multilingual BERT, a Small Corpus, and a Small Treebank
* Accepted to Findings of EMNLP 2020 
  Access Paper or Ask Questions

RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models


Sep 25, 2020
Samuel Gehman, Suchin Gururangan, Maarten Sap, Yejin Choi, Noah A. Smith

Figure 1 for RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models
Figure 2 for RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models
Figure 3 for RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models
Figure 4 for RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models
* Findings in EMNLP 2020 
  Access Paper or Ask Questions

Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation


Jun 28, 2020
Jungo Kasai, Nikolaos Pappas, Hao Peng, James Cross, Noah A. Smith

Figure 1 for Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation
Figure 2 for Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation
Figure 3 for Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation
Figure 4 for Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation
* Added references and acknowledgements 
  Access Paper or Ask Questions

A Mixture of $h-1$ Heads is Better than $h$ Heads


May 13, 2020
Hao Peng, Roy Schwartz, Dianqi Li, Noah A. Smith

Figure 1 for A Mixture of $h-1$ Heads is Better than $h$ Heads
Figure 2 for A Mixture of $h-1$ Heads is Better than $h$ Heads
Figure 3 for A Mixture of $h-1$ Heads is Better than $h$ Heads
Figure 4 for A Mixture of $h-1$ Heads is Better than $h$ Heads
* ACL2020 
  Access Paper or Ask Questions

The Right Tool for the Job: Matching Model and Instance Complexities


May 09, 2020
Roy Schwartz, Gabriel Stanovsky, Swabha Swayamdipta, Jesse Dodge, Noah A. Smith

Figure 1 for The Right Tool for the Job: Matching Model and Instance Complexities
Figure 2 for The Right Tool for the Job: Matching Model and Instance Complexities
Figure 3 for The Right Tool for the Job: Matching Model and Instance Complexities
Figure 4 for The Right Tool for the Job: Matching Model and Instance Complexities
* ACL 2020; 12 pages; code available in https://github.com/allenai/sledgehammer 
  Access Paper or Ask Questions

Don't Stop Pretraining: Adapt Language Models to Domains and Tasks


May 05, 2020
Suchin Gururangan, Ana Marasović, Swabha Swayamdipta, Kyle Lo, Iz Beltagy, Doug Downey, Noah A. Smith

Figure 1 for Don't Stop Pretraining: Adapt Language Models to Domains and Tasks
Figure 2 for Don't Stop Pretraining: Adapt Language Models to Domains and Tasks
Figure 3 for Don't Stop Pretraining: Adapt Language Models to Domains and Tasks
Figure 4 for Don't Stop Pretraining: Adapt Language Models to Domains and Tasks
* ACL 2020 
  Access Paper or Ask Questions