作者位次 #5/6ACL 2026Findings· Findings of ACL
W iki S eeker: Rethinking the Role of Vision-Language Models in Knowledge-Based Visual Question Answering
Yingjian Zhu · Xinming Wang · Kun Ding · ying wang · Bin Fan · Shiming Xiang
摘要
Multi-modal Retrieval-Augmented Generation (RAG) has emerged as a highly effective paradigm for Knowledge-Based Visual Question Answering (KB-VQA). Despite recent advancements, prevailing methods still primarily depend on images as the retrieval key, and often overlook or misplace the role of Vis...