ASMR-Bench: Auditing for Sabotage in ML Research

Computer Science > Artificial Intelligence [Submitted on 17 Apr 2026] ASMR-Bench: Auditing for Sabotage in ML Research Eric Gan, Aryan Bhatt, Buck Shlegeris, Julian Stastny, Vivek Hebbar As AI systems are increasingly used to conduct research autonomously, misaligned systems could introduce subtle flaws that produce misleading results while evading detection. We introduce ASMR-Bench (Auditing for Sabotage in ML Research), a benchmark for evaluating the ability of auditors to detect sabotage in ML research codebases. ASMR-Bench consists of 9 ML research codebases with sabotaged variants that produce qualitatively different experimental results. Each sabotage modifies implementation details, such as hyperparameters, training data, or evaluation code, while preserving the high-level methodology described in the paper. We evaluated frontier LLMs and LLM-assisted human auditors on ASMR-Bench and found that both struggled to reliably detect sabotage: the best performance was an AUROC of 0.77 and a top-1 fix rate of 42%, achieved by Gemini 3.1 Pro. We also tested LLMs as red teamers and found that LLM-generated sabotages were weaker than human-generated ones but still sometimes evaded same-capability LLM auditors. We release ASMR-Bench to support research on monitoring and auditing techniques for AI-conducted research. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2604.16286 [cs.AI]   (or arXiv:2604.16286v1 [cs.AI] for this version)   https://doi.org/10.48550/arXiv.2604.16286 Focus to learn more Submission history From: Eric Gan [view email] [v1] Fri, 17 Apr 2026 17:47:32 UTC (2,713 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev   |   next > new | recent | 2026-04 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)