作者位次 #5/11ACL 2026Findings· Findings of ACL
Rethinking Multiple-Choice Questions for RLVR : Unlocking Potential via Distractor Design
Xu Guo · Qiming Ge · Jian Tong · Kedi Chen · Jin Zhang · Xiaogui Yang · Xuan Gao · Haijun Lv · Zhihui Lu · Yicheng Zou · Qipeng Guo
摘要
Reinforcement Learning with Verifiable Rewards (RLVR) significantly enhances the reasoning capabilities of Large Language Models. When applied to RLVR, Multiple-Choice Questions (MCQs) offer a scalable source of verifiable data but risk inducing reward hacking, where models shortcut reasoning via...