2026
$G^2$-Reader: Dual Evolving Graphs for Multimodal Document QA
ICML 2026poster
Retrieval-augmented generation is a practical paradigm for question answering over long documents, but it remains brittle for multimodal reading where text, tables, and figures are interleaved across many pages. First, flat chunking breaks document-native structure and cross-modal alignment, yieldin…