作者位次 #6/12ACL 2026Poster· Natural Language Processing
Act as you think: Reinforcing Consistent Reasoning in Medical Visual Question Answering
Songtao Jiang · Yuan Wang · Ruizhe Chen · Yan Zhang · Ruilin Luo · Bohan Lei · Yeying Jin · Sibo Song · ZhiBo Yang · Jimeng Sun · JIAN Wu · Zuozhu Liu
摘要
While reinforcement learning from verifiable rewards (RLVR) has been proven highly effective for enhancing reasoning, its application to medical visual question answering (Med-VQA) is hampered by models producing reasoning inconsistent with either the visual evidence or the final answer. Our anal...