Takuto Asakura, Ph.D.

• Researcher at Stockmark
• Project Researcher at The University of Tokyo
• Email: takuto [at] is.s.u-tokyo.ac.jp
• Links: ORCID, Google Scholar, Researchmap, GitHub, LinkedIn, Keybase, Blog

Research Interest

• Natural Language Processing / Computational Linguistics
• Mathematical Language Processing / Mathematical Linguistics
• Mathematical Information Retrieval

Experience

• Researcher, Stockmark, December 2024―present
• Software Engineer, SB Intuitions, Tokyo, July 2024―November 2024
• Project Researcher, The University of Tokyo, Tokyo, April 2024―present
• Software Engineer, LY Corporation, Tokyo, October 2023―June 2024
• Software Engineer, LINE Corporation, Tokyo, April 2023―September 2023
• Part-time Software Engineer, LINE Corporation, Tokyo, May 2021―March 2023
• Research Assistant, The University of Tokyo, Tokyo, April 2021―March 2023
• Teaching Assistant (Academic Writing), The University of Tokyo, Tokyo, September 2020―March 2021
• Assistant Technical Staff, Research Organization of Information and Systems, Tokyo, April 2018―March 2021
• Teaching Assistant (Science), Seiko Gakuin High School, Yokohama, September 2017―March 2018

Publications (Refereed)

• Aamin Dev, Takuto Asakura, Rune Sætre. An Approach to Co-reference Resolution and Formula Grounding for Mathematical Identifiers using Large Language Models. In Proceedings of The 2nd Workshop on Mathematical Natural Language Processing (MathNLP 2024).
  [paper] [bib]
• Takuto Asakura, Yusuke Miyao. What Is Needed for Intra-document Disambiguation of Math Identifiers?. In Proceedings of The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024).
  [paper] [bib] [poster]
• Tarek Saier, Mayumi Ohta, Takuto Asakura, Michael Färber. HyperPIE: Hyperparameter Information Extraction from Scientific Publications. In Proceedings of 46th European Conference on Information Retrieval (ECIR 2024).
  [preprint] [paper] [bib]
• Takuto Asakura, Yusuke Miyao, Akiko Aizawa. Building Dataset for Grounding of Formulae — Annotating Coreference Relations Among Math Identifiers. In Proceedings of 13th Conference on Language Resources and Evaluation (LREC 2022). pp. 4851―4858, 2022.
  [paper] [bib] [slides] [video] [resource]
• Takuto Asakura, Yusuke Miyao, Akiko Aizawa, Michael Kohlhase. MioGatto: A Math Identifier-oriented Grounding Annotation Tool. In 13th MathUI Workshop at 14th Conference on Intelligent Computer Mathematics (MathUI 2021).
  [preprint] [paper] [slides] [code]
• Takuto Asakura, André Greiner-Petter, Akiko Aizawa, Yusuke Miyao. Towards Grounding of Formulae. In Proceedings of First Workshop on Scholarly Document Processing (SDP 2020). pp. 138―147, 2020.
  [paper] [bib] [poster] [resource]
• Takuto Asakura, Jin-Dong Kim, Yasunori Yamamoto, Yuka Tateisi, Toshihisa Takagi. A Quantitative Evaluation of Natural Language Question Interpretation for Question Answering Systems. In Proceedings of 8th Joint International Semantic Technology Conference (JIST 2018). pp. 215―231, 2018.
  [preprint] [paper] [bib] [slides] [code]

Publications (Non-refereed) and Talks

• Takuto Asakura, André Greiner-Petter, Akiko Aizawa, Yusuke Miyao. Dataset Creation for Grounding of Formulae. In SCIDOCA 2020.
  [slides] [resource]
• Takuto Asakura. The design concept for llmk—Light LaTeX Make. In TUG 2020.
  [paper] [slides] [video] [code]
• Takuto Asakura. A TeX-oriented research topic: Synthetic analysis on mathematical expressions and natural language. In TUG 2019.
  [paper] [slides]
• Takuto Asakura. Understanding Scientific Documents with Synthetic Analysis on Mathematical Expressions and Natural Language. In CICM 2019.
  [paper] [slides]
• Takuto Asakura. New Features in Texdoc 3.0. In BachoTeX 2018.
  [slides] [code]
• Takuto Asakura. Implementing bioinformatics algorithms in TeX―the Gotoh package, a case study. In TUG@BachoTeX 2017. (W.J. Martin Prize)
  [paper] [slides] [code]

Domestic Publications (Non-refereed) and Talks

• 朝倉卓人，宮尾祐介．数式識別子の文書内曖昧性の解消．言語処理学会第30回年次大会 (NLP 2024)．
  [paper] [slides]
• 朝倉卓人，宮尾祐介，相澤彰子．MioGattoによる数式グラウンディングデータセットの構築．言語処理学会第28回年次大会 (NLP 2022)．（委員特別賞）
  [paper] [slides]
• 朝倉卓人，宮尾祐介，相澤彰子．数式グラウンディングに向けて．NLP若手の会第16回シンポジウム (YANS 2021)．
  [poster] [code]
• 朝倉卓人，André Greiner-Petter，相澤彰子，宮尾祐介．数学概念への数式グラウンディングのためのデータセット．言語処理学会第26回年次大会 (NLP 2020)．（若手奨励賞）
  [paper] [poster]

Education

• Ph.D. in Computer Science, April 2020―March 2024
  • Department of Computer Science, The University of Tokyo, Japan
  • Major: Natural Language Processing
  • Supervisor: Prof. Yusuke Miyao
• Master of Informatics, April 2018―March 2020
  • Department of Informatics, SOKENDAI, Japan
  • Major: Natural Language Processing
  • Supervisor: Prof. Yusuke Miyao and Prof. Akiko Aizawa
• Bachelor of Science, April 2014―March 2018
  • Department of Bioinformatics and Systems Biology, The University of Tokyo, Japan
  • Major: Bioinformatics
  • Supervisor: Prof. Toshihisa Takagi

Awards and Honors

• 言語処理学会第28回年次大会 (NLP 2022) 委員特別賞
• 言語処理学会第27回年次大会 (NLP 2021) 功績賞
• 言語処理学会第26回年次大会 (NLP 2020) 若手奨励賞
• TUG@BachoTeX 2017 W.J. Martin Prize
• 日本生物学オリンピック2013 銀賞

Grants

• JST ACT-X Acceleration Phase, 5M JPY, April 2023―March 2024
• JSPS Research Fellowship (DC2), April 2022―March 2023
• JST ACT-X, 4.5M JPY, November 2020―March 2023
• TeX Development Fund, 1,000 USD, July 2018

Languages

• native: Japanese
• fluent: English
• basic: German, French, Esperanto

Computer Skills

• Programming Languages
  • fluent: Python, Ruby, Lua, TeX/LaTeX
  • basic: C, C++, TypeScript, OCaml, Rust, etc.
• Tools and Technologies: Unix-like systems, Git, SQL, PyTorch, Docker, Elasticsearch, etc.

Software

The followings are featured projects. The entire list is available from GitHub.

• MioGatto: an annotation tool for grounding of formulae
• Texdoc: documentation search tool for TeX Live
• llmk: a build tool for LaTeX documents