Professional Summary
I am currently a researcher at the Japan AI Safety Institute under the supervision of Prof. Satoshi Sekine, and a research associate at the University of Electro-Communications under Satoshi Hara. I received my Master’s degree from the same university under the supervision of Kei Harada. My research focuses on hallucination detection in vision language models and narrative understanding, especially temporal discourse understanding. I am currently seeking Ph.D. positions.
Education
MSc in Design Thinking and Data Science
The University of Electro-Communications
BSc in Design Thinking and Data Science
The University of Electro-Communications
Interests
Hallucination Detection in Vision–Language Models
Temporal Discourse Understanding in Language Models
Narrative Understanding
Recent Publications
(2026).
EnsemHalDet: Robust VLM Hallucination Detection via Ensemble of Internal State Detectors.
In ACL2026 SRW.
(2025).
BookAsSumQA: An Evaluation Framework for Aspect-Based Book Summarization via Question Answering.
In IJCNLP-AACL2025 SRW.
(2025).
Ensembling Multiple Hallucination Detectors Trained on VLLM Internal Representations.
In KDD Cup workshop 2025 at KDD2025.
Experience
Researcher
Japan AI Safety Institute
Research Intern
Ichikara Inc.
Research Intern (AI & Data Analytics Group)
DOCOMO Innovations, Inc.
Education
MSc in Design Thinking and Data Science
The University of Electro-Communications
BSc in Design Thinking and Data Science
The University of Electro-Communications
Awards
Student Award 2026 at University of Electro-Communications
2026-03
Second Place, Ichikara-Try 2025
2025-09
5th Place at KDD Cup 2025 Meta CRAG-MM Challenge
2025-08