三田 雅人

Principles Transfer, Parameters Don't: Universal Grammar-Inspired Multilingual Pre-Pretraining 

Proceedings of the BabyLM 2026. 2026年10月  査読有り筆頭著者

Language Acquisition Device in Large Language Models 

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics 2026年7月  査読有り筆頭著者

Developmentally-plausible Working Memory Shapes a Critical Period for Language Acquisition 

Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics 2025年7月  査読有り筆頭著者責任著者

Towards Automated Document Revision: Grammatical Error Correction, Fluency Edits, and Beyond. 

Proceedings of the 19th Workshop on Innovative Use of NLP for Building Educational Applications(BEA) 251-265 2024年  

Large Language Models Are State-of-the-Art Evaluator for Grammatical Error Correction. 

Proceedings of the 19th Workshop on Innovative Use of NLP for Building Educational Applications(BEA) 68-77 2024年  

Striking Gold in Advertising: Standardization and Exploration of Ad Text Generation. 

Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) 955-972 2024年  

Not Eliminate but Aggregate: Post-Hoc Control over Mixture-of-Experts to Address Shortcut Shifts in Natural Language Understanding. 

Transactions of the Association for Computational Linguistics 12 1268-1289 2024年  

Revisiting Meta-evaluation for Grammatical Error Correction. 

Transactions of the Association for Computational Linguistics 12 837-855 2024年  

Do Grammatical Error Correction Models Realize Grammatical Generalization? 

Findings of the Association for Computational Linguistics: ACL/IJCNLP 2021 4554-4561 2021年  

GitHub Typo Corpus: A Large-Scale Multilingual Dataset of Misspellings and Grammatical Errors. 

Proceedings of The 12th Language Resources and Evaluation Conference(LREC) 6761-6768 2020年  

A Self-Refinement Strategy for Noise Reduction in Grammatical Error Correction. 

Findings of the Association for Computational Linguistics: EMNLP 2020 267-280 2020年  

Encoder-Decoder Models Can Benefit from Pre-trained Masked Language Models in Grammatical Error Correction. 

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics(ACL) 4248-4254 2020年  

Cross-Corpora Evaluation and Analysis of Grammatical Error Correction Models - Is Single-Corpus Evaluation Enough? 

Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies 1309-1314 2019年  

An Empirical Study of Incorporating Pseudo Data into Grammatical Error Correction. 

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing 1236-1242 2019年