I work at the intersection of Human–AI Interaction and Natural Language Processing (NLP). My current focus is modeling temporal dynamics in multimodal dyadic interactions and building interpretable, human-aligned analysis tools.
- Human–AI Interaction, Social Computing
- Multimodal Interaction, temporal dynamics, synchrony
- Explainable AI, human-centered evaluation
- NLP / LLMs for interpretation and analysis
- Master’s thesis (JAIST): temporal patterns in multimodal behaviors during speed-dating conversations and their relationship to post-interaction impressions.
- Language Style Matching: reproducible pipeline for Japanese language style matching (LSM / rLSM / rolling-window rLSM).
- HAI 2025 (Workshop): “Analyzing the Relationship between Conversational Language Style and Outcomes of Speed-Dating.”
- Google Summer of Code (Rocket.Chat): contributor (2024), mentor (2025). I value transparent communication and written documentation to keep teams aligned.
- Email: yuriko_kikuchi[@]jaist.ac.jp
- LinkedIn: http://linkedin.com/in/yuriko-kikuchi



