Auditing LLM Benchmarks with Item Response Theory
Mirrored from arXiv — NLP / Computation & Language for archival readability. Support the source by reading on the original site.
Computer Science > Computation and Language
Title:Auditing LLM Benchmarks with Item Response Theory
Abstract:LLM benchmark labels are frozen at release and silently propagated into downstream benchmarks, errors and all. We introduce an Item Response Theory-based indicator that surfaces likely mislabels at 95% precision in the top 200 examples across seven preference and multiple-choice benchmarks using responses from 114 models, outperforming a supervised classifier. We trace these errors to mechanical labeling heuristics, upstream annotation mistakes inherited unchanged from source datasets, and fundamentally ambiguous items without a defensible single label. The same model fit reveals that reward models specialize in stylistic preference rather than factual knowledge, and identifies one frontier reward model that agrees with detected mislabels at 78% accuracy versus 38% for its peers, consistent with benchmark contamination or benchmark-specific over-optimization.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2605.30504 [cs.CL] |
| (or arXiv:2605.30504v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.30504
arXiv-issued DOI via DataCite (pending registration)
|
Access Paper:
- View PDF
- TeX Source
References & Citations
Bookmark
Bibliographic and Citation Tools
Code, Data and Media Associated with this Article
Demos
Recommenders and Search Tools
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
More from arXiv — NLP / Computation & Language
-
Protocol for evaluating ChatGPT in biomedical association generation and verification using a RAG-enabled, cross-model majority voting workflow
Jun 1
-
Exploring Autonomous Agentic Data Engineering for Model Specialization
Jun 1
-
Domain Adaptation and Reasoning Frameworks in Language Models: A Controlled Experiment with Historical Cosmology
Jun 1
-
Cross-Lingual Steering for Figurative Language Generation
Jun 1
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.