ConferenceAnnual Meeting of the Association for Computational Linguistics2025
Stress Testing Large Reasoning Models
Zhuoshi Pan, Qizhi Pei, Yu Li, et al.
Introduces REST, a stress-testing framework that evaluates large reasoning models by presenting multiple problems simultaneously, revealing performance degradation and stronger discriminative power than single-question benchmarks.
15Jul 14, 2025ReasoningBenchmarks
arXiv