About Me
I am a 3rd-year PhD student at the University of Tokyo. My research explores multi-agent reinforcement learning, computational social science, and language models for understanding human behavior.
Research Interests
- Multi-agent Reinforcement Learning — Cooperation in a decentralized manner.
- Computational Social Science — Applying machine learning to social science.
- LLM for Behavior Understanding — Replicating human behavior and utilizing common sense reasoning.
News
- Aug. 2026
Two first-author papers have been accepted to the EMNLP 2026 Main Conference (acceptance rate 15.4%).
- Jul. 2026
Selected for the 2026 KAKUSEI Project with JPY 3 million in funding.
- Mar. 2026
Received Young Researcher Award (21/517 top 4%) at NLP 2026.
- Mar. 2026
Our paper on Masked Diffusion Language Models has been selected for oral presentation at NLP 2026 (32 out of 799 submissions).
- Feb. 2026
Release new preprint on Masked Diffusion Language Models.
- Sep. 2025
Our paper on LLM’s self-reflection has been accepted to NeurIPS 2025.
Publications
International Conference (Peer-reviewed)
-
Cross-Centaur: Modeling Individual Decision-Making Across Psychological Tasks with Language Models
The 2026 Conference on Empirical Methods in Natural Language Processing · acceptance rate 15.4%
-
MobQA: A Benchmark Dataset for Semantic Understanding of Human Mobility Data through Question Answering
The 2026 Conference on Empirical Methods in Natural Language Processing · acceptance rate 15.4%
-
Self Iterative Label Refinement via Robust Unlabeled Learning
The Thirty-Ninth Annual Conference on Neural Information Processing Systems · poster, acceptance rate 24.5%
-
Text2Traj2Text: Learning-by-Synthesis Framework for Contextual Captioning of Human Movement Trajectories
International Natural Language Generation Conference · long paper
-
Quantitative and Qualitative Analyses on Child's Animate Tendency
Cognitive feelings: Towards multi-disciplinary approaches for realizing artificial systems with cognitive capacities
-
Rashomon Bound Regression: a Reproducible and Explorable Method for Regression-Based Analysis
International Conference on Computational Social Science · poster
-
Counterfactual Fairness Filter for Fair-Delay Multi-Robot Navigation
International Conference on Autonomous Agents and Multi-Agent Systems · oral, acceptance rate 23.3%
Preprints
-
Where-to-Unmask: Ground-Truth-Guided Unmasking Order Learning for Masked Diffusion Language Models
arXiv preprint
-
Sequential Bayesian experimental designs via reinforcement learning
arXiv preprint
Education
- D.Sc in Graduate School of Arts and Sciences, The University of Tokyo (2024.4-present)
- M.S in Graduate School of Interdisciplinary Information Studies, The University of Tokyo (2022.4-2024.3)
- B.E in Systems Innovation, Faculty of Engineering, The University of Tokyo (2019.4-2022.3)
Job Experiences
- Sakana AI, Research Intern (2026.4-present)
- OMRON SINIC X Corporation, Research Intern (2025.12-2026.3)
- CyberAgent AI Lab, Part-time Researcher (2023.10-2025.10)
- OMRON SINIC X Corporation, Research Intern (2022.3-2023.8)
- OMRON SINIC X Corporation, Research Intern (2020.8-2020.9)
Awards and Honors
Scholarship and Grant
KAKUSEI Project by AIST, 3,000,000 yen (salary and research expenses), 2026. Research RA in Value Exchange Engineering by RIISE@UTokyo, 2,000,000 yen, 2023-2024. The foundation Matsuo Scholarship Society, Scholarship and commuting expenses grant, free accommodation in student dormitories, provision of meals (free), health insurance expenses grant, total amount granted: approximately 5,000,000 yen, 2019-2023.
Academic Award
Young Researcher Award (21/517), The 32nd Annual Meeting of the Association for Natural Language Processing@Japan, 2026. Hatakeyama Prize, Tomakomai National Institute of Technology, 2019.
Academic Services
Reviewer
ACL Jan, 2026