田邊光（たなべひかる）と申します。電気通信大学大学院（UEC）の博士前期課程に所属し、Vision & Language に関する研究や Web サービス開発に取り組んでいます。最近は特定ドメインにおける MLLMs の活用に興味があります。
Specializing in Vision-and-Language and web development. Developed methods leveraging Multimodal Large Language Models (MLLMs) and volume estimation to improve energy estimation from food images. Designed and implemented evaluation pipelines for LLMs to enhance performance analysis, drawing from my experience in ML engineering. Actively seeking opportunities to contribute to projects involving LLMs or MLLMs.
