RAGOCR renders retrieved documents as images and beats naive RAG by 15% on one-eighth the input tokens.

arXiv 2608.00765 compresses retrieved docs into query-conditioned visual representations, sidestepping the trade-off where hard compression is query-aware but weak and soft compression is strong but needs costly offline encoding. Beats both baselines across varying retrieval depths. The mechanism transfers without the visual encoder: a query-aware dynamic resolution pass that renders highly relevant passages at higher granularity while aggressively downsampling peripheral documents. Relevance-proportional budget allocation is something any RAG pipeline can adopt today.