I am currently a Research Manager at Stockmark Inc., working on research in natural language processing. Email: hello@whiro.me

Current Work

AI for Innovation

I work on “AI for Innovation,” especially AI for discovering new business opportunities from manufacturing technologies.
I am also involved in launching our technology-application discovery business.
While leading a joint research project with AIST, I develop generative AI technologies that support idea generation and application discovery.

  • Wataru Hirota, Chung-Chi Chen, Tomoko Ohkuma, Tomoki Taniguchi, Tatsuya Ishigaki. Overview of PBIG Shared Task at AgentScen 2025: Product Business Idea Generation from Patents. ACL Anthology.
  • Keisuke Ueda, Wataru Hirota, Takuto Asakura, Takahiro Omi, Kosuke Takahashi, Kosuke Arima, Tatsuya Ishigaki. Exploring Design of Multi-Agent LLM Dialogues for Research Ideation. SIGDIAL 2025. arXiv.
  • A new article about this project is published

Knowledge Graphs

I conduct research on automatically building knowledge graphs as a knowledge base for LLMs to perform application discovery and ideation.
Alongside improving core relation-extraction accuracy, I am developing technology for automatically constructing knowledge graphs from documents owned by clients.
I am also collaborating with Prof. Kentaro Inui (Tohoku University / MBZUAI) on knowledge graph construction research (link).

Past Activities

Megagon Labs

As a Research Associate at Megagon Labs in Mountain View, CA (US), I worked on multilingual embeddings, entity linking for database records, and conversational AI.

  • Wataru Hirota, Yoshihiko Suhara, Behzad Golshan, Wang-Chiew Tan. Semantic Cross-lingual Sentence Embeddings. RepL4NLP (hosted by ACL). 2019.
  • Wataru Hirota, Yoshihiko Suhara, Behzad Golshan, Wang-Chiew Tan. Emu: Enhancing Multilingual Sentence Embeddings with Semantic Specialization. AAAI. 2020. arXiv.
  • Yuliang Li, Jinfeng Li, Yoshihiko Suhara, Jin Wang, Wataru Hirota, Wang-Chiew Tan. Deep Entity Matching: Challenges and Opportunities. Journal of Data and Information Quality. 2021. ACM.
  • Jin Wang, Yuliang Li, Wataru Hirota. Machamp: A Generalized Entity Matching Benchmark. CIKM 2021 (Resource Track). arXiv.

University / Graduate School

During university and graduate school, I conducted research in bioinformatics and information retrieval.

  • Mitsuhiro Eto, Wataru Hirota, Shigeto Seno, Hideo Matsuda. Asymmetric Integration of Single-Cell Transcriptomic Data using Latent Dirichlet Allocation and Procrustes Analysis. IEEE BIBM. 2018. IEEE Xplore.