Takeshi OnishiAI/NLP Research Scientist

Publications and talks

• 牧野 寛也, 大西 健史, 大滝 啓介, 堺 浩之. 未解決問題の判別・生成能力を測る LLM 評価ベンチマーク. 第4回Science of Science 研究会, 2026
• 大滝 啓介, 牧野 寛也, 大西 健史, 堺 浩之. OpenAlexとLLMを用いた論文特徴の分解と可視化. 第4回Science of Science 研究会, 2026
• 大西 健史. JOB HUNTING. 第74回SCJSF&JABAフォーラム, 2025年
• Takeshi Onishi, Davy Weissenbacher, Matthew Scotch, Graciela Gonzalez-Hernandez. ToponymQA: Toponym Linking as multiple-choice QA by Retrieval-Augmented Generation. The outstanding poster presentation at The First Japan-US Science Forum in Southern California "Frontiers in Science: Innovations in AI and Data Science", February 2025.
• Ari Z. Klein, Tirthankar Dasgupta, Lauren Gryboski, Sudeshna Jana, Sedigh Khademi, Guillermo Lopez-Garcia, Diego Mazzotti, Takeshi Onishi, Jeanne Powell, Lisa Raithel, Swati Rajwal, Roland Roller, Abeed Sarker, Manjira Sinha, Philippe Thomas, Elena Tutubalina, Dongfang Xu, Pierre Zweigenbaum, Graciela Gonzalez-Hernandez. Overview of the 10th Social Media Mining for Health (#SMM4H) and Health Real-World Data (HeaRD) Shared Tasks at ICWSM 2025. In the SMM4H-HeaRD 2025 workshop at the AAAI International Conference on Web and Social Media (ICWSM 2025), June 2025.
• Ari Z. Klein, Davy Weissenbacher, Karen O'Connor, Amir Elyaderani, Ivan Flores Amaro, Takeshi Onishi, Su Golder, Kaelen Spiegel, Matthew Scotch, Graciela Gonzalez-Hernandez. Detection of patient metadata in published articles for genomic epidemiology using machine learning and large language models. medRxiv [Preprint], April 2025.
• Takeshi Onishi. Applied LLMs: Classification of Patient Metadata sentences. In an oral session of Computational Biomedicine Health Language Processing Annual Symposium 2024
• Takeshi Onishi and James Caverlee. Political Bias of Large Language Models in Few-shot News Summarization. Advances in Bias and Fairness in Information Retrieval, October 2024.
• Kazuhiko Shinoda, Takeshi Onishi, Masashi Sugiyama. Noisy-labeled Domain Adaptation under Generalized Label Shift. In proceedings of The 37th Annual Conference of the Japanese Society for Artificial Intelligence, June 2023.
• Takeshi Onishi. What is a Ph.D. program? In a special lecture at Toyota Technological Institute, April 2023.
• Takeshi Onishi. Various academic/industrial research environments in the U.S./Japan. In a lab talk at Intelligent Information Media Lab, September 2022.
• Takeshi Onishi. Alumni interview. In the annual school magazine of Toyota Technological Institute, July 2022.
• Takeshi Onishi. What is a Ph.D. program? In a special lecture at Toyota Technological Institute, May 2022.
• Takeshi Onishi, Tomoya Takatani, Hirotaka Kaji, Masashi Sugiyama. Missing value imputation on heterogeneous data by an autoencoder. In an oral session of the 24th Information-based Induction Sciences, November 2021.
• Takeshi Onishi. How to make your research career. In a special lecture at Gifu University, January 2021.
• Takeshi Onishi. How does AI understand texts? -From the basics to my thesis-. In a public talk of JRCC, December 2020.
• Takeshi Onishi. Relation/Entity-Centric Reading Comprehension. Ph.D. Thesis, August 2020.
• Takeshi Onishi, Takuya Kadohira, Ikumu Watanabe. Relation extraction from scientific articles for material informatics. In an invited talk of 20th International Union of Materials Research Societies International Conference in Asia, September 2019.
• Takeshi Onishi, Davy Weissenbacher, Ari Klein, Karen O’Connor, and Graciela Gonzalez-Hernandez. Dealing with Medication Non-Adherence Expressions in Twitter. In Proceedings of the 2018 EMNLP Workshop SMM4H: The 3rd Social Media Mining for Health Applications Workshop & Shared Task, October 2018.
• Takeshi Onishi, Takuya Kadohira, Ikumu Watanabe. Relation extraction with weakly supervised learning based on process-structure-property-performance reciprocity. Science and Technology of Advanced Materials, September 2018.
• Takeshi Onishi, Hai Wang, Kevin Gimpel, David McAllester. Emergent Predication Structure in Hidden State Vectors of Neural Readers. A Best Paper in Repl4NLP: 2nd Workshop on Representation Learning for NLP, August 2017.
• Takeshi Onishi. Material Development by Artificial Intelligence -another application of the data science-. In a special lecture at Toyota Technological Institute, December 2016.
• Takeshi Onishi, Hai Wang, Mohit Bansal, Kevin Gimpel, and David McAllester. Who did What: A large-scale person-centered cloze dataset. In proceedings of Empirical Methods in Natural Language Processing, November 2016.
• Takeshi Onishi, Takuya Kadohira, Ikumu Watanabe. Causal Relation Extraction from Natural Language Texts for Material Development. In a poster session of NIMS WEEK, October 2016.
• Takeshi Onishi. How to apply Artificial Intelligence and Bigdata to your field. In a public talk of JRCC, October 2015.

Patent

• 杉山 将, 篠田 和彦, 大西 健史. 学習方法、学習装置、及びプログラム. 国立大学法人東京大学, トヨタ自動車株式会社. 特願2023-062528, 2023年4月出願 (公開).
• 光田 英司, 大西 健史, 西口 浩司, 中川 あすか, 酒井 瑞樹. 情報処理装置、情報処理方法およびプログラム. トヨタ自動車株式会社. 特願2022-148221, 2022年9月出願 (特許登録).
• 大西 健史, 杉山 将. 欠損値推定方法、機械学習方法および欠損値推定装置. トヨタ自動車株式会社, 国立大学法人東京大学. 特願2021-165372, 2021年10月出願 (特許登録).