2026/8/14 15:25:06 阿里:揭示LLM科学推理中的捷径作弊
📖标题:Right Answer, Wrong Method: Shortcut Hacking Misleads the Evaluation of LLM Reasoning on Frontier Science Benchmarks
🌐来源:arXiv, 2608.02442v1
🛎️文章简介
🔸研究问题:仅凭…