Momose Oyama

• I am a first-year Ph.D student at Shimodaira Lab, Kyoto University.
• Research interest:
  • Understanding How Embedding Geometry Encodes Meaning
  • Interpretation of Neural Network Models
• E-Mail: oyama.momose [atmark] sys.i.kyoto-u.ac.jp
• Google Scholar

Publications

1. Momose Oyama, Hiroaki Yamagiwa, Hidetoshi Shimodaira.
   Understanding Higher-Order Correlations Among Semantic Components in Embeddings. EMNLP 2024 (Main).
   [arXiv]
2. Hiroaki Yamagiwa*, Momose Oyama*, Hidetoshi Shimodaira.
   Discovering Universal Geometry in Embeddings with ICA. EMNLP 2023 (Main).
   [arXiv] [ACL Anthology]
3. Momose Oyama, Sho Yokoi, Hidetoshi Shimodaira.
   Norm of Word Embedding Encodes Information Gain. EMNLP 2023 (Main).
   [arXiv] [ACL Anthology]

Preprint

1. Hiroaki Yamagiwa, Momose Oyama, Hidetoshi Shimodaira.
   Revisiting Cosine Similarity via Normalized ICA-transformed Embeddings.
   [arXiv]

Awards

• Student Outstanding Presentation Award at IBIS2023
• Incentive award at YANS2023
• Finalist for the Presentation Award at IBIS2022
• Young Researcher Award at NLP2022

Experiences

• 2023/06 - Present Research Assistant, RIKEN Center for Advanced Intelligence Project
• 2022/04 - 2023/05 Research Assistant, Kyoto University
• 2021/07 - 2022/03 Research Assistant, RIKEN Center for Advanced Intelligence Project

Education

• 2024/04 - Present Ph.D in Informatics, Kyoto University.
• 2022/04 - 2024/03 M.S. in Informatics, Kyoto University.
• 2018/04 - 2022/03 B.E. in Informatics, Kyoto University.