作者位次 #5/9ACL 2026Findings· Findings of ACL
Beyond Reasoning Gains: Mitigating General-Capability Forgetting in Large Reasoning Models
Hoang Phan · Xianjun Yang · Yuanshun Yao · Jingyu Zhang · Shengjie Bi · Xiaocheng Tang · Madian Khabsa · Lijuan Liu · Deren Lei
摘要
Reinforcement learning with verifiable rewards (RLVR) has delivered impressive gains in mathematical and multimodal reasoning and has become a standard post-training paradigm for contemporary language and vision-language models. However, the RLVR recipe introduces a significant risk of capability...