← Back to all news
arXiv AI By Xinming Tu (Minta), Tianze Wang (Minta), Yingzhou (Minta), Lu, Kexin Huang, Yuanhao Qu, Sara Mostafavi

Who Guards the Benchmarks? Automated Auditing of LLM Agent Benchmarks

Read the original on arXiv AI →

The paper introduces BenchGuard, a framework that uses advanced language models to audit execution-based agent benchmarks. BenchGuard systematically cross-verifies benchmark artifacts through structured LLM protocols, optionally incorporating agent solutions or execution traces for deeper diagnostics. In tests on ScienceAgentBench and BIXBench Verified-50, BenchGuard uncovered numerous author-confirmed and expert-identified issues, including fatal errors that made tasks unsolvable, and demonstrated cost-effective auditing for complex bioinformatics tasks.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv AI
Aug 28

Beyond Execution: Auditing Experimental Fidelity in LLM-Driven Scientific Research

The paper "Beyond Execution: Auditing Experimental Fidelity in LLM-Driven Scientific Research" argues that large language model agents must faithfully implement reference methods, design experiments that truly test claims, and provide supporting evidence. It reports that agents often hallucinate methodology—reducing datasets, substituting components, or drawing conclusions from limited resources—leading to false claims. To counter this, the authors introduce ABE‑Ralph, a reference‑anchored auditing framework that structures experimental constraints, guides implementation, and verifies results, achieving a 93% robust execution rate across 30 reproduction runs and matching or exceeding state‑of‑the‑art performance on 5 NatureBench tasks. "whyItMatters":"The study demonstrates that evaluating AI scientists requires more than code execution; it must ensure experimental design and evidence truly support the claimed scientific outcomes."

By Lezhi Yu, Xiaogang Xu, Yuhua Zhou, Shuibing He, Aimin Pan
llmsagentsbenchmarkssafety
More like this →
arXiv AI
Aug 10

An End-to-End Agent Auditing Engine

arXiv:2608. 07346v1 Announce Type: new Abstract: With the rapid advancement of large language models (LLMs), harnesses have become essential infrastructure for deploying agents across a wide range of domains.

By Haoning Wang, Mingxun Zhang, Chenyue Yu, Yingjun Shang, Xia Hu, Guanchu Wang, Na Zou
llmsagents
More like this →
arXiv AI
Aug 11

$A^2E$ : An End-to-End Agent Auditing Engine

arXiv:2608. 07346v2 Announce Type: replace Abstract: With the rapid advancement of large language models (LLMs), harnesses have become essential infrastructure for deploying agents across a wide range of domains.

By Haoning Wang, Mingxun Zhang, Chenyue Yu, Yingjun Shang, Xia Hu, Guanchu Wang, Na Zou
llmsagents
More like this →
arXiv AI
Jul 14

FIRE-Bench: Evaluating AI Agents on the Rediscovery of Scientific Insights

arXiv:2602. 02905v2 Announce Type: replace Abstract: Autonomous agents powered by large language models (LLMs) promise to accelerate scientific discovery end-to-end, but rigorously evaluating their capacity for verifiable discovery remains a central challenge.

By Zhen Wang, Fan Bai, Zhongyan Luo, Jinyan Su, Kaiser Sun, Xinle Yu, Jieyuan Liu, Kun Zhou, Claire Cardie, Mark Dredze, Zhiting Hu, Eric P. Xing
llmsagentsbenchmarks
More like this →