GTBench evaluates LLMs as math research assistants in graph theory

Arxiv CS.AI (top papers)

GTBench is a curriculum-grounded benchmark testing LLMs as mathematical reasoning assistants in graph theory. It provides structured tasks to assess reliability and problem-solving capabilities.

benchmarks

Arxiv CS.AI (top papers)·Noujoud NaderIbrahem AljabeaPatrick DiehlDeepti Gupta

8 days ago