Mr. Yuya Tahara (Arai)(4th year student, Humanics)participated in the LA-Bench 2025 AI competition for experimental procedure generation as part of the team "cabbage patch" and won the Grand Prize in the LLM Evaluation category.
Mr. Yuya Tahara (Arai)(4th year student, Humanics)participated in the LA-Bench 2025 AI competition for experimental procedure generation as part of the team "cabbage patch" and won the Grand Prize in the LLM Evaluation category.
Title : LA-Bench 2025: Experimental Procedure Generation AI Competition
Team： cabbage patch
Award： Grand prize in the LLM evaluation category
Achieved a score of 9.289, and was officially recognized as an award-winning team.
Competition outline:：LA-Bench 2025 evaluates AI systems that generate practical and executable experimental procedures based on inputs such as:
Experiment objectives
Materials used
Original protocols
Expected final outcomes
Supplementary Presentation：
A presentation and award ceremony were held at:
The 40th Annual Conference of the Japanese Society for Artificial Intelligence (JSAI 2026)
Special Session: KS-21 AI and Competitions
[Message from Mr. Yuya Tahara (Arai) ]
Through LA-Bench 2025, I worked on a challenge that connects laboratory automation and large language models to generate executable experimental procedures from natural language instructions.
The Humanics programme provides opportunities to work across biology, informatics, AI, robotics, and related fields.
I encourage future Humanics students to take on interdisciplinary challenges and explore research topics beyond the boundaries of their original expertise.
