RAG, Fine-Tuning & ContextManuel Faysse et al. (Illuin Technology, CentraleSupélec and others)
ColPali published
ColPali published — embeds document page images directly with a VLM and retrieves via late interaction
Why it matters
A lineage of visual document retrieval that skips the OCR and parsing pipeline
Source — Primary sources first; secondary reports are labelled “Press”
Paperarxiv.orghttps://arxiv.org/abs/2407.01449Added in Week 35, 2026 · See that week