Regular biography
Hitomi Yanaka is a faculty member at The University of Tokyo, Department of CS, with expertise in Computational Linguistics, Inference, and Natural Language Processing. Her work focuses on advancing understanding and applications in these areas. Yanaka's research interests align with her academic profile, which is hosted on her personal website. She is affiliated with The University of Tokyo and contributes to the field through her scholarly activities.
Scholar-generated biography
Hitomi Yanaka is a researcher at The University of Tokyo and RIKEN, specializing in Natural Language Processing and Semantics. Her work explores the capabilities and limitations of neural models in understanding linguistic phenomena such as monotonicity reasoning, systematicity, and compositional semantics. Yanaka has developed datasets like HELP and Jamp to evaluate model generalization and bias, focusing on Japanese language tasks. Her research also includes multimodal logical inference, grammatical error correction, and the development of open-source Japanese large language models. Her contributions span both theoretical and applied aspects of NLP, emphasizing the intersection of formal semantics and machine learning.