Research Interests

  • Natural Language Processing (NLP)
  • Large Language Model (LLM)
  • Hallucination
  • Paraphrase

Publications

International Conference

  • Sora Kadotani, Kosuke Nishida, Kyosuke Nishida. Learning from Hallucinations: Mitigating Hallucinations in LLMs via Internal Representation Intervention. In Findings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (IJCNLP-AACL 2025). December 2025. (Acceptance rate 30%) [Paper] [Slides] [Poster]
  • Sora Kadotani, Yuki Arase. Monolingual Phrase Alignment as Parse Forest Mapping. In Proceedings of the 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023). July 2023. (Acceptance rate 47%) [Paper] [Slides]
  • Sora Kadotani, Tomoyuki Kajiwara, Yuki Arase, Makoto Onizuka. Edit Distance Based Curriculum Learning for Paraphrase Generation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing: Student Research Workshop (ACL-IJCNLP 2021 SRW). August 2021. (Acceptance rate 39%) [Paper]

Domestic Conference

  • 門谷 宙, 西田 光甫, 西田 京介, 齋藤 邦子. In-model anti-expertによる大規模言語モデルのハルシネーション抑制. NLP若手の会19回シンポジウム (YANS 2024). September 2024. [Poster]
  • 門谷 宙, 西田 光甫, 西田 京介, 齋藤 邦子. 大規模言語モデルは自身のHallucinationを検知できるか?. 言語処理学会第30回年次大会 (NLP 2024). March 2024. [Paper] [Slides]
  • 門谷 宙, 荒瀬 由紀. 構文森を用いたフレーズアラインメント. NLP若手の会第17回シンポジウム (YANS 2022). August 2022.
  • 門谷 宙, 梶原 智之, 荒瀬 由紀, 鬼塚 真. 編集距離に基づくカリキュラム学習を用いたスタイル変換. 人工知能学会第35回全国大会 (JSAI 2021). June 2021. [Paper]

Preprint

  • Sora Kadotani, Kosuke Nishida, Kyosuke Nishida. Can LLMs Detect Their Own Hallucinations?. [Paper]

Talks

  • 門谷宙. 国際会議ACL2023参加報告. 第258回自然言語処理研究発表会 (第258回 NL研). December 2023. [Slides]

Experiences

  • 2023-04 ~
    NTT Human Informatics Laboratories
  • 2022-01 ~ 2022-02
    NTT Human Informatics Laboratories (Internship)
  • 2021-09
    Yahoo Japan Corporation (Internship)

Education

  • 2023-03
    Master of Information Science, Graduate School of Information Science and Technology, Onizuka Lab.
  • 2021-03
    Bachelor of Engineering, Osaka University, Onizuka Lab.