CLIMB: Confidence-Guided Complementary Evidence for Multimodal Retrieval-Augmented Generation
Multimodal large language models (MLLMs) have shown strong visual reasoning abilities, but knowledge-intensive visual question answering often requires external textual evidence beyond the image and the model's parametric knowledge. Existing multimodal RAG systems commonly rely on Top-$K$ retrieval or reranking, which...
Hang Gao, Wu-Jiang Xu, Zhixing Zhang et al.
· 0 citations