Who Guards the Benchmarks? Automated Auditing of LLM Agent Benchmarks
Xinming Tu · Tianze Wang · Yingzhou Lu · Kexin Huang · Yuanhao Qu · Sara Mostafavi
We introduce BenchGuard, an automated auditing framework that uses frontier LLMs to detect hidden specification errors and evaluation flaws in execution-based AI agent benchmarks.
摘要
As benchmarks grow in complexity, many apparent agent failures are not failures of the agent at all---they are failures of the benchmark itself: broken specifications, implicit assumptions, and rigid evaluation scripts that penalize valid alternative approaches. We propose employing frontier LLMs as systematic auditors of evaluation infrastructure, and realize this vision through **benchguard**, the first framework explicitly designed for joint cross-artifact auditing of execution-based agent benchmarks. **benchguard** cross-verifies all benchmark artifacts via structured LLM protocols, optionally incorporating agent solutions or execution traces as additional diagnostic evidence. Deployed on two prominent scientific benchmarks, **benchguard** identified 12 author-confirmed issues in ScienceAgentBench---including fatal errors rendering tasks unsolvable---and exactly matched 83.3\% of expert-identified issues on the BIXBench Verified-50 subset, catching defects that prior human review missed entirely. A full audit of 50 complex bioinformatics tasks costs under \$15, making automated benchmark auditing a practical and valuable complement to human review. A preliminary native-format audit of ProgramBench further demonstrates cross-format applicability. These findings point toward AI-assisted benchmark development, where frontier models serve not only as subjects of evaluation but as active participants in validating the evaluation infrastructure itself.