Canasai Kruengkrai カナサイ・クルンカライ

I am a research scientist at the RIKEN Guardian Robot Project (GRP).
My recent research focuses on the reliability of language models under data limitations, imperfect inputs, and interactive environments.

Publications

• What Changes in Hidden States When Model Merging Hurts Task Performance?
  Canasai Kruengkrai and Koichiro Yoshino
  To appear in AACL-IJCNLP 2026
  [code] [checkpoints]
• Teaching Text Agents to Learn Sequential Decision Making from Failure
  Canasai Kruengkrai and Koichiro Yoshino
  Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL), pages 31619–31635, 2025
  [code] [checkpoints] [PEFT extension] [PEFT checkpoints]
• Revisiting Pathologies of Neural Models under Input Reduction
  Canasai Kruengkrai and Junichi Yamagishi
  Findings of the Association for Computational Linguistics: ACL 2023, pages 11504–11517, 2023
  [code] [checkpoints]
• Mitigating the Diminishing Effect of Elastic Weight Consolidation
  Canasai Kruengkrai and Junichi Yamagishi
  Proceedings of the 29th International Conference on Computational Linguistics (COLING), pages 4568–4574, 2022
  [code] [checkpoints]
• A Multi-Level Attention Model for Evidence-Based Fact Checking
  Canasai Kruengkrai, Junichi Yamagishi, and Xin Wang
  Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 2447–2460, 2021
  [code] [checkpoints]
• Improving Low-Resource Named Entity Recognition using Joint Sentence and Token Labeling
  Canasai Kruengkrai, Thien Hai Nguyen, Sharifah Mahani Aljunied, and Lidong Bing
  Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL), pp. 5898–5905, 2020
• Learning to Flip the Sentiment of Reviews from Non-Parallel Corpora
  Canasai Kruengkrai
  Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pp. 6312–6317, 2019
  [code]
• Better Exploiting Latent Variables in Text Modeling
  Canasai Kruengkrai
  Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL), pp. 5527–5532, 2019
  [code]

See all publications