GTBench: A Curriculum-Grounded Benchmark for Evaluating LLMs as Mathematical Research Assistants in Graph Theory 文章

ArXiv CS.AI2026-06-03NEWSen作者: Noujoud Nader, Ibrahem Aljabea, Patrick Diehl, Deepti Gupta
大语言模型
查看原文 →
关系图谱

GTBench: A Curriculum-Grounded Benchmark for Evaluating LLMs as Mathematical Research Assistants in Graph Theory · 相关技术

相关技术

Large language modelLLM-as-judge evaluationexact-match evaluationzero-shot promptingLLM-as-judgeChain-of-Thought promptingzero-shotgraph theory