Progressive Multimodal Search and Reasoning for Knowledge-Intensive Visual Question Answering

Changin Choi, Wonseok Lee, Jungmin Ko +1

Knowledge-intensive visual question answering (VQA) requires external knowledge beyond image content, demanding precise visual grounding and coherent integration of visual and text…