| Are we modeling the task or the annotator? an investigation of annotator bias in natural language understanding datasets M Geva, Y Goldberg, J Berant EMNLP 2019, 2019 | 260 | 2019 |
| Beyond the imitation game: Quantifying and extrapolating the capabilities of language models A Srivastava, A Rastogi, A Rao, AAM Shoeb, A Abid, A Fisch, AR Brown, ... arXiv preprint arXiv:2206.04615, 2022 | 180 | 2022 |
| Injecting Numerical Reasoning Skills into Language Models M Geva, A Gupta, J Berant ACL 2020, 2020 | 144 | 2020 |
| Break It Down: A Question Understanding Benchmark T Wolfson, M Geva, A Gupta, M Gardner, Y Goldberg, D Deutch, J Berant TACL, 2020 | 128 | 2020 |
| Did aristotle use a laptop? a question answering benchmark with implicit reasoning strategies M Geva, D Khashabi, E Segal, T Khot, D Roth, J Berant Transactions of the Association for Computational Linguistics 9, 346-361, 2021 | 123 | 2021 |
| Transformer feed-forward layers are key-value memories M Geva, R Schuster, J Berant, O Levy arXiv preprint arXiv:2012.14913, 2020 | 114 | 2020 |
| DiscoFuse: A Large-Scale Dataset for Discourse-based Sentence Fusion M Geva, E Malmi, I Szpektor, J Berant NAACL-HLT 2019 1, 3443-3455, 2019 | 42 | 2019 |
| Transformer feed-forward layers build predictions by promoting concepts in the vocabulary space M Geva, A Caciularu, KR Wang, Y Goldberg arXiv preprint arXiv:2203.14680, 2022 | 34 | 2022 |
| Emergence of communication in an interactive world with consistent speakers B Bogin, M Geva, J Berant arXiv preprint arXiv:1809.00549, 2018 | 34 | 2018 |
| Scrolls: Standardized comparison over long language sequences U Shaham, E Segal, M Ivgi, A Efrat, O Yoran, A Haviv, A Gupta, W Xiong, ... arXiv preprint arXiv:2201.03533, 2022 | 22 | 2022 |
| Evaluating semantic parsing against a simple web-based question answering model A Talmor, M Geva, J Berant *SEM 2018, 2017 | 16 | 2017 |
| Don't Blame the Annotator: Bias Already Starts in the Annotation Instructions M Parmar, S Mishra, M Geva, C Baral arXiv preprint arXiv:2205.00415, 2022 | 15 | 2022 |
| Learning to Search in Long Documents Using Document Structure M Geva, J Berant COLING 2018, 2018 | 14 | 2018 |
| Lm-debugger: An interactive tool for inspection and intervention in transformer-based language models M Geva, A Caciularu, G Dar, P Roit, S Sadde, M Shlain, B Tamir, ... arXiv preprint arXiv:2204.12130, 2022 | 12 | 2022 |
| Break, perturb, build: Automatic perturbation of reasoning paths through question decomposition M Geva, T Wolfson, J Berant Transactions of the Association for Computational Linguistics 10, 111-126, 2022 | 10 | 2022 |
| What's in your Head? Emergent Behaviour in Multi-Task Transformer Models M Geva, U Katz, A Ben-Arie, J Berant arXiv preprint arXiv:2104.06129, 2021 | 8 | 2021 |
| Analyzing transformers in embedding space G Dar, M Geva, A Gupta, J Berant arXiv preprint arXiv:2209.02535, 2022 | 7 | 2022 |
| Crawling the internal knowledge-base of language models R Cohen, M Geva, J Berant, A Globerson arXiv preprint arXiv:2301.12810, 2023 | 5 | 2023 |
| Jump to Conclusions: Short-Cutting Transformers With Linear Transformations A Yom Din, T Karidi, L Choshen, M Geva arXiv e-prints, arXiv: 2303.09435, 2023 | 3* | 2023 |
| Inferring implicit relations in complex questions with language models U Katz, M Geva, J Berant Findings of the Association for Computational Linguistics: EMNLP 2022, 2548-2566, 2022 | 3* | 2022 |
Jonathan BerantAssociate Professor, Tel-Aviv UniversityVerified email at cs.tau.ac.il
