作者位次 #2/5ACL 2026Findings· Findings of ACL
Faithful-First Reasoning, Planning, and Acting for Multimodal LLM s
Junxian Li · Xinyue Xu · Sai Ma · Di Zhang · Sichao Li
摘要
Multimodal Large Language Models (MLLMs) frequently suffer from unfaithfulness, generating reasoning chains that drift from visual evidence or contradict final predictions. We propose Faithful-First Reasoning, Planning, and Acting (RPA) framework in which FaithEvi provides step-wise and chain-lev...