Computing Answers to Complex Questions in Broad Domains

Informazioni relative al progetto

DELPHI

ID dell’accordo di sovvenzione: 802800

DOI

10.3030/802800

Progetto chiuso

Data della firma CE 14 Novembre 2018

Data di avvio 1 Aprile 2019

Data di completamento 31 Marzo 2026

Finanziato da

• EXCELLENT SCIENCE - European Research Council (ERC)

Costo totale

€ 1 499 375,00

Contributo UE

€ 1 499 375,00

Coordinato da

TEL AVIV UNIVERSITY
Israel

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Pubblicazioni

Conference proceedings (31)

Diverse Demonstrations Improve In-context Compositional Generalization

Autori: Itay Levy, Ben Bogin, Jonathan Berant
Pubblicato in: Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2023, Pagina/e 1401-1422
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2023.acl-long.78

MAKING RETRIEVAL-AUGMENTED LANGUAGEMODELS ROBUST TO IRRELEVANT CONTEXT

Autori: Ori Yoran Tomer Wolfson Ori Ram Jonathan Berant
Pubblicato in: Numero Published as a conference paper at ICLR 2024, 2024
Editore: The Twelfth International Conference on Learning Representations ICLR 2024

Unobserved Local Structures Make Compositional Generalization Hard

Autori: Ben Bogin, Shivanshu Gupta, Jonathan Berant
Pubblicato in: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, 2026, Pagina/e 2731-2747
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2022.emnlp-main.175

"""COVR: A test-bed for Visually Grounded Compositional Generalization with real images"""

Autori: Ben Bogin, Shivanshu Gupta, Matt Gardner, Jonathan Berant
Pubblicato in: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2021, Pagina/e 9824–9846
Editore: Association for Computational Linguistics

MultimodalQA: Complex Question Answering over Text, Tables and Images

Autori: Alon Talmor, Ori Yoran, Amnon Catav, Dan Lahav, Yizhong Wang, Akari Asai, Gabriel Ilharco, Hannaneh Hajishirzi, Jonathan Berant
Pubblicato in: Proceedings of the 2021 International Conference on Learning Representations, 2021
Editore: ICLR 2021

A Simple and Effective Model for Answering Multi-span Questions

Autori: Elad Segal, Avia Efrat, Mor Shoham, Amir Globerson, Jonathan Berant
Pubblicato in: Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2020, Pagina/e 3074-3080
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2020.emnlp-main.248

What’s in Your Head? Emergent Behaviour in Multi-Task Transformer Models

Autori: Mor Geva, Uri Katz, Aviv Ben-Arie, Jonathan Berant
Pubblicato in: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2021
Editore: Association for Computational Linguistics

Achieving Model Robustness Through Discrete Adversarial Training

Autori: Maor Ivgi and Jonathan Berant
Pubblicato in: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2021
Editore: ACL 2021

Learning To Retrieve Prompts for In-Context Learning

Autori: Ohad Rubin, Jonathan Herzig, Jonathan Berant
Pubblicato in: Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2022, Pagina/e 2655-2671
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2022.naacl-main.191

Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors

Autori: Ido Amos, Jonathan Berant, Ankit Gupta
Pubblicato in: ICLR 2024 oral, 2024
Editore: Published as a conference paper at ICLR 2024

Scaling Laws Under the Microscope: Predicting Transformer Performance from Small Scale Experiments

Autori: Maor Ivgi, Yair Carmon, Jonathan Berant
Pubblicato in: Findings of the Association for Computational Linguistics: EMNLP 2022, 2026, Pagina/e 7354-7371
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2022.findings-emnlp.544

Inferring Implicit Relations in Complex Questions with Language Models

Autori: Uri Katz, Mor Geva, Jonathan Berant
Pubblicato in: Findings of the Association for Computational Linguistics: EMNLP 2022, 2026, Pagina/e 2548-2566
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2022.findings-emnlp.188

Finding needles in a haystack: Sampling Structurally-diverse Training Sets from Synthetic Data for Compositional Generalization

Autori: Inbar Oren, Jonathan Herzig, Jonathan Berant
Pubblicato in: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2021
Editore: Association for Computational Linguistics

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

Autori: Ori Yoran, Alon Talmor, Jonathan Berant
Pubblicato in: Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2022, Pagina/e 6016-6031
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2022.acl-long.416

Training Vision-Language Models with Less Bimodal Supervision

Autori: Elad Segal, Ben Bogin, Jonathan Berant
Pubblicato in: 2022
Editore: AKBC 2022
DOI: 10.48550/arxiv.2211.00262

Injecting Numerical Reasoning Skills into Language Models

Autori: Mor Geva, Ankit Gupta, Jonathan Berant
Pubblicato in: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020, Pagina/e 946-958
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2020.acl-main.89

SCROLLS: Standardized CompaRison Over Long Language Sequences

Autori: Uri Shaham, Elad Segal, Maor Ivgi, Avia Efrat, Ori Yoran, Adi Haviv, Ankit Gupta, Wenhan Xiong, Mor Geva, Jonathan Berant, Omer Levy
Pubblicato in: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, 2026, Pagina/e 12007-12021
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2022.emnlp-main.823

Scene Graph tO Image Generation with Contextualized Object Layout Refinement

Autori: Maor Ivgi, Yaniv Benny, Avichai Ben-David, Jonathan Berant, Lior Wolf
Pubblicato in: 2021 IEEE International Conference on Image Processing (ICIP), 2021, Pagina/e 2428-2432, ISBN 978-1-6654-4115-5
Editore: IEEE
DOI: 10.1109/icip42928.2021.9506651

Leap-Of-Thought: Teaching Pre-Trained Models to Systematically Reason Over Implicit Knowledge

Autori: Talmor, Alon; Tafjord, Oyvind; Clark, Peter; Goldberg, Yoav; Berant, Jonathan
Pubblicato in: Advances in Neural Information Processing Systems 33 pre-proceedings (NeurIPS 2020), Numero 4, 2020
Editore: NeurIPS 2020

CommonsenseQA 2.0: Exposing the Limits of AI through Gamification

Autori: Alon Talmor, Ori Yoran, Ronan Le Bras, Chandra Bhagavatula, Yoav Goldberg, Yejin Choi, Jonathan Berant
Pubblicato in: Proceedings of the Thirty-fifth Conference on Neural Information Processing Systems (Datasets and Benchmarks Track), 2021
Editore: NeurIPS 2021

Improving Compositional Generalization in Semantic Parsing

Autori: Inbar Oren, Jonathan Herzig, Nitish Gupta, Matt Gardner, Jonathan Berant
Pubblicato in: Findings of the Association for Computational Linguistics: EMNLP 2020, 2020, Pagina/e 2482-2495
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2020.findings-emnlp.225

SmBoP: Semi-autoregressive Bottom-up Semantic Parsing

Autori: Ohad Rubin, Jonathan Berant
Pubblicato in: Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2021, Pagina/e 311-324
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2021.naacl-main.29

In-Context Learning with Long-Context Models: An In-Depth Exploration

Autori: Amanda Bertsch, Maor Ivgi, Emily Xiao, Uri Alon, Jonathan Berant, Matthew R. Gormley, Graham Neubig
Pubblicato in: Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), 2025, Pagina/e 12119-12149
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2025.naacl-long.605

Evaluating the Evaluation of Diversity in Natural Language Generation

Autori: Guy Tevet, Jonathan Berant
Pubblicato in: Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, 2021, Pagina/e 326-346
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2021.eacl-main.25

ASSISTANTBENCH: Can Web Agents SolveRealistic and Time-Consuming Tasks?

Autori: Ori Yoran Samuel Joseph Amouyal Chaitanya Malaviya Ben Bogin Ofir Press Jonathan Berant
Pubblicato in: 2024
Editore: arXiv:2407.15711v2 [cs.CL] 21 Oct 2024

Evaluating Models’ Local Decision Boundaries via Contrast Sets

Autori: Matt Gardner, Yoav Artzi, Victoria Basmov, Jonathan Berant, Ben Bogin, Sihao Chen, Pradeep Dasigi, Dheeru Dua, Yanai Elazar, Ananth Gottumukkala, Nitish Gupta, Hannaneh 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
Pubblicato in: Findings of the Association for Computational Linguistics: EMNLP 2020, 2020, Pagina/e 1307-1323
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2020.findings-emnlp.117

Span-based Semantic Parsing for Compositional Generalization

Autori: Jonathan Herzig, Jonathan Berant
Pubblicato in: Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), 2021, Pagina/e 908-921
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2021.acl-long.74

Differentiable Scene Graphs

Autori: Moshiko Raboh, Roei Herzig, Jonathan Berant, Gal Chechik, Amir Globerson
Pubblicato in: 2020 IEEE Winter Conference on Applications of Computer Vision (WACV), 2020, Pagina/e 1477-1486, ISBN 978-1-7281-6553-0
Editore: IEEE
DOI: 10.1109/wacv45572.2020.9093297

Obtaining Faithful Interpretations from Compositional Neural Networks

Autori: Sanjay Subramanian, Ben Bogin, Nitish Gupta, Tomer Wolfson, Sameer Singh, Jonathan Berant, Matt Gardner
Pubblicato in: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020, Pagina/e 5594-5608
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2020.acl-main.495

Weakly Supervised Text-to-SQL Parsing through Question Decomposition

Autori: Tomer Wolfson, Daniel Deutch, Jonathan Berant
Pubblicato in: Findings of the Association for Computational Linguistics: NAACL 2022, 2022, Pagina/e 2528-2542
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2022.findings-naacl.193

Answering Questions by Meta-Reasoning over Multiple Chains of Thought

Autori: Ori Yoran, Tomer Wolfson, Ben Bogin, Uri Katz, Daniel Deutch, Jonathan Berant
Pubblicato in: Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, 2023, Pagina/e 5942-5966
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2023.emnlp-main.364

Other (3)

Question Decomposition with Dependency Graphs

Autori: Matan Hasson, Jonathan Berant
Pubblicato in: AKBC 2021, 2021
Editore: non-archival

Efficient Long-Text Understanding with Short-Text Models

Autori: Maor Ivgi, Uri Shaham, Jonathan Berant
Pubblicato in: arXiv:2208.00748v3 [cs.CL] 27 Dec 2022, 2022
Editore: Accepted for publication in Transactions of the Association for Computational Linguistics (TACL), 2023.
DOI: 10.48550/arxiv.2208.00748

Simplifying and Understanding State Space Modelswith Diagonal Linear RNNs

Autori: Ankit Gupta, Harsh Mehta, Jonathan Berant
Pubblicato in: 2022
Editore: OTHER
DOI: 10.48550/arxiv.2212.00768

Peer reviewed articles (6)

Retrieval-Pretrained Transformer: Long-range Language Modeling with Self-retrieval

Autori: Ohad Rubin, Jonathan Berant
Pubblicato in: Transactions of the Association for Computational Linguistics, Numero 12, 2024, Pagina/e 1197-1213, ISSN 2307-387X
Editore: MIT Press
DOI: 10.1162/tacl_a_00693

oLMpics-On What Language Model Pre-training Captures

Autori: Alon Talmor, Yanai Elazar, Yoav Goldberg, Jonathan Berant
Pubblicato in: Transactions of the Association for Computational Linguistics, Numero 8, 2020, Pagina/e 743-758, ISSN 2307-387X
Editore: The MIT Press
DOI: 10.1162/tacl_a_00342

Break It Down: A Question Understanding Benchmark

Autori: Tomer Wolfson; Mor Geva; Ankit Gupta; Matt Gardner; Yoav Goldberg; Daniel Deutch; Jonathan Berant
Pubblicato in: Transactions of the Association for Computational Linguistics, Vol 8, Pp 183-198 (2020), Numero 2, 2020, ISSN 2307-387X
Editore: MIT Press Journals
DOI: 10.1162/tacl_a_00309

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

Autori: Mor Geva, Tomer Wolfson, Jonathan Berant
Pubblicato in: Transactions of the Association for Computational Linguistics, 2021, 2021, ISSN 2307-387X
Editore: MIT Press

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

Autori: Mor Geva, Daniel Khashabi, Elad Segal, Tushar Khot, Dan Roth, Jonathan Berant
Pubblicato in: Transactions of the Association for Computational Linguistics, Numero 9, 2021, Pagina/e 346-361, ISSN 2307-387X
Editore: MIT Press
DOI: 10.1162/tacl_a_00370

Latent compositional representations improve systematic generalization in grounded question answering.

Autori: Ben Bogin, Sanjay Subramanian, Matt Gardner, Jonathan Berant
Pubblicato in: Transactions of the Association for Computational Linguistics (TACL) - volume 9, 2021, ISSN 2307-387X
Editore: MIT Press