About me

I am Atsumoto Ohashi, a Ph.D. candidate in Informatics at Nagoya University, with a primary research focus on dialogue systems. My work has centered on optimizing task-oriented dialogue systems through the integration of NLP and reinforcement learning (RL) techniques. Recently, my research interests have expanded to spoken language models and full-duplex spoken dialogue systems for real-time and more natural conversational experiences.

My CV can be found here (PDF).

Education

  • April 2023 – Present
    Ph.D. student in Informatics
    Graduate School of Informatics, Nagoya University

  • April 2021 – March 2023
    Master of Informatics
    Graduate School of Informatics, Nagoya University

  • April 2017 – March 2021
    Bachelor of Informatics
    School of Informatics, Nagoya University

Research Experience

  • September 2025 – Present
    Kyutai, Paris, France
    Visiting PhD Student

  • June 2025 – September 2025
    National Institute of Informatics (NII), Remote
    Research Assistant
    Project: Development of spoken dialogue models at LLM-jp project

  • August 2023 – September 2023
    CyberAgent, Inc., Tokyo, Japan
    Research Intern
    Project: Analyzing minimum bayes risk (MBR) decoding for neural machine translation models

Selected Publications

(* Equal contribution)

International Conferences (Refereed)

  • Atsumoto Ohashi, Shinya Iizuka, Jingjing Jiang, and Ryuichiro Higashinaka.
    “Towards a Japanese Full-duplex Spoken Dialogue System”
    The 26th Interspeech Conference (INTERSPEECH 2025)
    [Paper] [Website] [Code]

  • Atsumoto Ohashi and Ryuichiro Higashinaka.
    “Universal Post-Processing Networks for Joint Optimization of Modules in Task-Oriented Dialogue Systems”
    The 39th Annual AAAI Conference on Artificial Intelligence (AAAI 2025)
    [Paper] [Code]

  • Atsumoto Ohashi, Ukyo Honda, Tetsuro Morimura, and Yuu Jinnai.
    “On the True Distribution Approximation of Minimum Bayes-Risk Decoding”
    The 2024 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2024)
    [Paper] [Code]

  • Atsumoto Ohashi*, Ryu Hirai*, Shinya Iizuka, and Ryuichiro Higashinaka.
    “JMultiWOZ: A Large-Scale Japanese Multi-Domain Task-Oriented Dialogue Dataset”
    The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
    [Paper] [Code]

  • Atsumoto Ohashi and Ryuichiro Higashinaka.
    “Enhancing Task-oriented Dialogue Systems with Generative Post-processing Networks”
    The 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP 2023)
    [Paper] [Code]

  • Atsumoto Ohashi and Ryuichiro Higashinaka.
    “Adaptive Natural Language Generation for Task-oriented Dialogue via Reinforcement Learning”
    The 29th International Conference on Computational Linguistics (COLING 2022)
    [Paper] [Code] [Blog (in Japanese)]

  • Atsumoto Ohashi and Ryuichiro Higashinaka.
    “Post-processing Networks: A Method for Optimizing Pipeline Task-oriented Dialogue Systems using Reinforcement Learning”
    The 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL 2022)
    [Paper] [Code] [Blog (in Japanese)]

Honors and Awards

  • Committee Special Award (32/726)
    The 31st Annual Meeting of the Association of Natural Language Processing (NLP), 2025
  • Young Researcher Encouragement Award (18/406)
    The 29th Annual Meeting of the Association of Natural Language Processing (NLP), 2023
  • Sponsorship Award (5/386)
    The 28th Annual Meeting of the Association of Natural Language Processing (NLP), 2022
  • Committee Special Award (16/365)
    The 28th Annual Meeting of the Association of Natural Language Processing (NLP), 2022
  • Outstanding Academic Achievement
    Department of Computer Science, Nagoya University, 2018
  • Chinese Speech Contest 4th prize
    Nanjing University Cup, 2017

Invited Talks

  • Minimum Bayes-Risk Decodingにおける性能変動の理解に向けて
    第59回 NLPコロキウム, 2024
    [Link] [Slide]

  • 国際会議参加報告 EMNLP 2023
    第99回 人工知能学会 言語・音声理解と対話処理研究会 (SLUD) 「第14回対話システムシンポジウム」, 2023
    [Link] [Slide]

  • 国際会議参加報告 SIGDIAL2022/YRRSDS2022
    第96回 人工知能学会 言語・音声理解と対話処理研究会 (SLUD) 「第13回対話システムシンポジウム」, 2022
    [Link] [Slide]

Scholarships and Grants

  • JSPS Research Fellowship for Young Scientists (DC2) [Link]
    April 2025 – March 2027
    • Research scholarship: 200,000 yen / month
    • Research Expenses: 800,000 yen / year
  • Nagoya University Interdisciplinary Frontier Fellowship [Link]
    April 2023 – March 2026
    • Research scholarship: 180,000 yen / month
    • Research Expenses: 250,000 yen / year

Experience

  • Workshop Organizer for YRRSDS 2024
    June 2024 – September 2024

  • Teaching Assistant for Intelligent Interface 2, Nagoya University
    June 2022 – August 2022

  • Short-term Study Abroad at Nanjing University, China
    February 2018 – March 2018