1
Auditing LLM Benchmarks with Item Response Theory
用项目反应理论深度剖析LLM基准测试,揭示评估偏差与模型能力真实度量。
arXiv:2605.30504v1 Announce Type: new Abstract: LLM benchmark labels are frozen at release and silently propagated into downstream benchmarks, errors …
用项目反应理论深度剖析LLM基准测试,揭示评估偏差与模型能力真实度量。
arXiv:2605.30504v1 Announce Type: new Abstract: LLM benchmark labels are frozen at release and silently propagated into downstream benchmarks, errors …