Rethinking RL Evaluation: Can Benchmarks Truly Reveal Failures of RL Methods? 事件

Rethinking RL Evaluation: Can Benchmarks Truly Reveal Failures of RL Methods? arXiv:2510.10541v2 Announce Type: replace-cross Abstract: Current benchmarks are inadequate for evaluating progress in reinforcement learning (RL) for large language models (LLMs).Despite recent benchmark gains reported for RL, we find that training on these benchmarks' training sets achieves nearly the same performance as training directly on the test sets, suggesting that the benchmarks cannot reliably separate furt