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 Jonathan Berant

Jonathan Berant

Break, Perturb, Build: Automatic Perturbation of Reasoning Paths through Question Decomposition


Jul 29, 2021
Mor Geva, Tomer Wolfson, Jonathan Berant

Figure 1 for Break, Perturb, Build: Automatic Perturbation of Reasoning Paths through Question Decomposition
Figure 2 for Break, Perturb, Build: Automatic Perturbation of Reasoning Paths through Question Decomposition
Figure 3 for Break, Perturb, Build: Automatic Perturbation of Reasoning Paths through Question Decomposition
Figure 4 for Break, Perturb, Build: Automatic Perturbation of Reasoning Paths through Question Decomposition

  Access Paper or Ask Questions

Turning Tables: Generating Examples from Semi-structured Tables for Endowing Language Models with Reasoning Skills


Jul 15, 2021
Ori Yoran, Alon Talmor, Jonathan Berant

Figure 1 for Turning Tables: Generating Examples from Semi-structured Tables for Endowing Language Models with Reasoning Skills
Figure 2 for Turning Tables: Generating Examples from Semi-structured Tables for Endowing Language Models with Reasoning Skills
Figure 3 for Turning Tables: Generating Examples from Semi-structured Tables for Endowing Language Models with Reasoning Skills
Figure 4 for Turning Tables: Generating Examples from Semi-structured Tables for Endowing Language Models with Reasoning Skills

  Access Paper or Ask Questions

Memory-efficient Transformers via Top-$k$ Attention


Jun 13, 2021
Ankit Gupta, Guy Dar, Shaya Goodman, David Ciprut, Jonathan Berant

Figure 1 for Memory-efficient Transformers via Top-$k$ Attention
Figure 2 for Memory-efficient Transformers via Top-$k$ Attention
Figure 3 for Memory-efficient Transformers via Top-$k$ Attention
Figure 4 for Memory-efficient Transformers via Top-$k$ Attention

  Access Paper or Ask Questions

Question Decomposition with Dependency Graphs


Apr 17, 2021
Matan Hasson, Jonathan Berant

Figure 1 for Question Decomposition with Dependency Graphs
Figure 2 for Question Decomposition with Dependency Graphs
Figure 3 for Question Decomposition with Dependency Graphs
Figure 4 for Question Decomposition with Dependency Graphs

  Access Paper or Ask Questions

What's in your Head? Emergent Behaviour in Multi-Task Transformer Models


Apr 13, 2021
Mor Geva, Uri Katz, Aviv Ben-Arie, Jonathan Berant

Figure 1 for What's in your Head? Emergent Behaviour in Multi-Task Transformer Models
Figure 2 for What's in your Head? Emergent Behaviour in Multi-Task Transformer Models
Figure 3 for What's in your Head? Emergent Behaviour in Multi-Task Transformer Models
Figure 4 for What's in your Head? Emergent Behaviour in Multi-Task Transformer Models

  Access Paper or Ask Questions

MultiModalQA: Complex Question Answering over Text, Tables and Images


Apr 13, 2021
Alon Talmor, Ori Yoran, Amnon Catav, Dan Lahav, Yizhong Wang, Akari Asai, Gabriel Ilharco, Hannaneh Hajishirzi, Jonathan Berant

Figure 1 for MultiModalQA: Complex Question Answering over Text, Tables and Images
Figure 2 for MultiModalQA: Complex Question Answering over Text, Tables and Images
Figure 3 for MultiModalQA: Complex Question Answering over Text, Tables and Images
Figure 4 for MultiModalQA: Complex Question Answering over Text, Tables and Images
* ICLR 2021 
  Access Paper or Ask Questions

Achieving Model Robustness through Discrete Adversarial Training


Apr 11, 2021
Maor Ivgi, Jonathan Berant

Figure 1 for Achieving Model Robustness through Discrete Adversarial Training
Figure 2 for Achieving Model Robustness through Discrete Adversarial Training
Figure 3 for Achieving Model Robustness through Discrete Adversarial Training
Figure 4 for Achieving Model Robustness through Discrete Adversarial Training

  Access Paper or Ask Questions

Value-aware Approximate Attention


Mar 17, 2021
Ankit Gupta, Jonathan Berant

Figure 1 for Value-aware Approximate Attention
Figure 2 for Value-aware Approximate Attention
Figure 3 for Value-aware Approximate Attention
Figure 4 for Value-aware Approximate Attention

  Access Paper or Ask Questions

BERTese: Learning to Speak to BERT


Mar 11, 2021
Adi Haviv, Jonathan Berant, Amir Globerson

Figure 1 for BERTese: Learning to Speak to BERT
Figure 2 for BERTese: Learning to Speak to BERT
Figure 3 for BERTese: Learning to Speak to BERT
Figure 4 for BERTese: Learning to Speak to BERT
* Accepted to EACL 2021 
  Access Paper or Ask Questions

Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategies


Jan 06, 2021
Mor Geva, Daniel Khashabi, Elad Segal, Tushar Khot, Dan Roth, Jonathan Berant

Figure 1 for Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategies
Figure 2 for Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategies
Figure 3 for Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategies
Figure 4 for Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategies
* Accepted for publication in Transactions of the Association for Computational Linguistics (TACL), 2021. Author's final version 
  Access Paper or Ask Questions

Few-Shot Question Answering by Pretraining Span Selection


Jan 02, 2021
Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy

Figure 1 for Few-Shot Question Answering by Pretraining Span Selection
Figure 2 for Few-Shot Question Answering by Pretraining Span Selection
Figure 3 for Few-Shot Question Answering by Pretraining Span Selection
Figure 4 for Few-Shot Question Answering by Pretraining Span Selection

  Access Paper or Ask Questions

Transformer Feed-Forward Layers Are Key-Value Memories


Dec 29, 2020
Mor Geva, Roei Schuster, Jonathan Berant, Omer Levy

Figure 1 for Transformer Feed-Forward Layers Are Key-Value Memories
Figure 2 for Transformer Feed-Forward Layers Are Key-Value Memories
Figure 3 for Transformer Feed-Forward Layers Are Key-Value Memories
Figure 4 for Transformer Feed-Forward Layers Are Key-Value Memories

  Access Paper or Ask Questions

SmBoP: Semi-autoregressive Bottom-up Semantic Parsing


Oct 23, 2020
Ohad Rubin, Jonathan Berant

Figure 1 for SmBoP: Semi-autoregressive Bottom-up Semantic Parsing
Figure 2 for SmBoP: Semi-autoregressive Bottom-up Semantic Parsing
Figure 3 for SmBoP: Semi-autoregressive Bottom-up Semantic Parsing
Figure 4 for SmBoP: Semi-autoregressive Bottom-up Semantic Parsing

  Access Paper or Ask Questions

Improving Compositional Generalization in Semantic Parsing


Oct 12, 2020
Inbar Oren, Jonathan Herzig, Nitish Gupta, Matt Gardner, Jonathan Berant

Figure 1 for Improving Compositional Generalization in Semantic Parsing
Figure 2 for Improving Compositional Generalization in Semantic Parsing
Figure 3 for Improving Compositional Generalization in Semantic Parsing
Figure 4 for Improving Compositional Generalization in Semantic Parsing

  Access Paper or Ask Questions

Learning Object Detection from Captions via Textual Scene Attributes


Sep 30, 2020
Achiya Jerbi, Roei Herzig, Jonathan Berant, Gal Chechik, Amir Globerson

Figure 1 for Learning Object Detection from Captions via Textual Scene Attributes
Figure 2 for Learning Object Detection from Captions via Textual Scene Attributes
Figure 3 for Learning Object Detection from Captions via Textual Scene Attributes
Figure 4 for Learning Object Detection from Captions via Textual Scene Attributes

  Access Paper or Ask Questions

Scene Graph to Image Generation with Contextualized Object Layout Refinement


Sep 24, 2020
Maor Ivgi, Yaniv Benny, Avichai Ben-David, Jonathan Berant, Lior Wolf

Figure 1 for Scene Graph to Image Generation with Contextualized Object Layout Refinement
Figure 2 for Scene Graph to Image Generation with Contextualized Object Layout Refinement
Figure 3 for Scene Graph to Image Generation with Contextualized Object Layout Refinement
Figure 4 for Scene Graph to Image Generation with Contextualized Object Layout Refinement

  Access Paper or Ask Questions

Span-based Semantic Parsing for Compositional Generalization


Sep 13, 2020
Jonathan Herzig, Jonathan Berant

Figure 1 for Span-based Semantic Parsing for Compositional Generalization
Figure 2 for Span-based Semantic Parsing for Compositional Generalization
Figure 3 for Span-based Semantic Parsing for Compositional Generalization
Figure 4 for Span-based Semantic Parsing for Compositional Generalization

  Access Paper or Ask Questions

A Simple Global Neural Discourse Parser


Sep 08, 2020
Yichu Zhou, Omri Koshorek, Vivek Srikumar, Jonathan Berant

Figure 1 for A Simple Global Neural Discourse Parser
Figure 2 for A Simple Global Neural Discourse Parser
Figure 3 for A Simple Global Neural Discourse Parser
Figure 4 for A Simple Global Neural Discourse Parser

  Access Paper or Ask Questions

Latent Compositional Representations Improve Systematic Generalization in Grounded Question Answering


Jul 01, 2020
Ben Bogin, Sanjay Subramanian, Matt Gardner, Jonathan Berant

Figure 1 for Latent Compositional Representations Improve Systematic Generalization in Grounded Question Answering
Figure 2 for Latent Compositional Representations Improve Systematic Generalization in Grounded Question Answering
Figure 3 for Latent Compositional Representations Improve Systematic Generalization in Grounded Question Answering
Figure 4 for Latent Compositional Representations Improve Systematic Generalization in Grounded Question Answering

  Access Paper or Ask Questions

Teaching Pre-Trained Models to Systematically Reason Over Implicit Knowledge


Jun 19, 2020
Alon Talmor, Oyvind Tafjord, Peter Clark, Yoav Goldberg, Jonathan Berant

Figure 1 for Teaching Pre-Trained Models to Systematically Reason Over Implicit Knowledge
Figure 2 for Teaching Pre-Trained Models to Systematically Reason Over Implicit Knowledge
Figure 3 for Teaching Pre-Trained Models to Systematically Reason Over Implicit Knowledge
Figure 4 for Teaching Pre-Trained Models to Systematically Reason Over Implicit Knowledge

  Access Paper or Ask Questions

GMAT: Global Memory Augmentation for Transformers


Jun 05, 2020
Ankit Gupta, Jonathan Berant

Figure 1 for GMAT: Global Memory Augmentation for Transformers
Figure 2 for GMAT: Global Memory Augmentation for Transformers
Figure 3 for GMAT: Global Memory Augmentation for Transformers
Figure 4 for GMAT: Global Memory Augmentation for Transformers

  Access Paper or Ask Questions

Obtaining Faithful Interpretations from Compositional Neural Networks


May 02, 2020
Sanjay Subramanian, Ben Bogin, Nitish Gupta, Tomer Wolfson, Sameer Singh, Jonathan Berant, Matt Gardner

Figure 1 for Obtaining Faithful Interpretations from Compositional Neural Networks
Figure 2 for Obtaining Faithful Interpretations from Compositional Neural Networks
Figure 3 for Obtaining Faithful Interpretations from Compositional Neural Networks
Figure 4 for Obtaining Faithful Interpretations from Compositional Neural Networks
* ACL 2020; first three authors contributed equally 
  Access Paper or Ask Questions

Evaluating the Evaluation of Diversity in Natural Language Generation


Apr 26, 2020
Guy Tevet, Jonathan Berant

Figure 1 for Evaluating the Evaluation of Diversity in Natural Language Generation
Figure 2 for Evaluating the Evaluation of Diversity in Natural Language Generation
Figure 3 for Evaluating the Evaluation of Diversity in Natural Language Generation
Figure 4 for Evaluating the Evaluation of Diversity in Natural Language Generation

  Access Paper or Ask Questions

Explaining Question Answering Models through Text Generation


Apr 12, 2020
Veronica Latcinnik, Jonathan Berant

Figure 1 for Explaining Question Answering Models through Text Generation
Figure 2 for Explaining Question Answering Models through Text Generation
Figure 3 for Explaining Question Answering Models through Text Generation
Figure 4 for Explaining Question Answering Models through Text Generation

  Access Paper or Ask Questions

Injecting Numerical Reasoning Skills into Language Models


Apr 09, 2020
Mor Geva, Ankit Gupta, Jonathan Berant

Figure 1 for Injecting Numerical Reasoning Skills into Language Models
Figure 2 for Injecting Numerical Reasoning Skills into Language Models
Figure 3 for Injecting Numerical Reasoning Skills into Language Models
Figure 4 for Injecting Numerical Reasoning Skills into Language Models
* ACL 2020 
  Access Paper or Ask Questions

Evaluating NLP Models via Contrast Sets


Apr 06, 2020
Matt Gardner, Yoav Artzi, Victoria Basmova, Jonathan Berant, Ben Bogin, Sihao Chen, Pradeep Dasigi, Dheeru Dua, Yanai Elazar, Ananth Gottumukkala, Nitish Gupta, Hanna Hajishirzi, Gabriel Ilharco, Daniel Khashabi, Kevin Lin, Jiangming Liu, Nelson F. Liu, Phoebe Mulcaire, Qiang Ning, Sameer Singh, Noah A. Smith, Sanjay Subramanian, Reut Tsarfaty, Eric Wallace, Ally Zhang, Ben Zhou

Figure 1 for Evaluating NLP Models via Contrast Sets
Figure 2 for Evaluating NLP Models via Contrast Sets
Figure 3 for Evaluating NLP Models via Contrast Sets
Figure 4 for Evaluating NLP Models via Contrast Sets

  Access Paper or Ask Questions