ViDoRe (Product): news timeline & CVEs

Inverting Multi-Vector Visual Document Indices

Researchers invert multi-vector visual document indices, recovering 47% of words and 45% of sensitive tokens on ViDoRe v3.

The paper shows multi-vector visual document retrievers, which store about a thousand patch vectors per page, can be inverted to reconstruct a page from its index alone. On ViDoRe v3, inversions from raw indices recover 47% of words and 45% of sensitive tokens, and as queries rank the source page first 98.4% of the time. Token pooling and shuffling cut word recall to about 8%, but a model that restores shuffled order raises first-rank recovery from 3.8% to 93.5%. Applied unchanged to another multi-vector retriever, inverted pages still rank their source first 70.2% of the time.

arXiv cs.AI / cs.LG / cs.CLOct 7, 12:10 UTC · 1d agoAI safety & security 2 sources1

Inverting Multi-Vector Visual Document Indices

Stored multi-vector document indices can be inverted to recover page words and sensitive tokens.

The paper shows multi-vector visual document retrievers, which store about a thousand patch vectors per page, can be inverted to reconstruct page images. Inversion is framed as conditional document image generation and requires the encoder, page shape, and vector order. On ViDoRe v3, inversions from raw indices recover 47% of words and 45% of sensitive tokens, and rank the source page first 98.4% of the time. Pooling and shuffling cut word recall to about 8%, but an order-restoration model raises shuffled first-rank matches from 3.8% to 93.5%; the same attack on another retriever still ranks the source first 70.2% of the time.

Hugging Face daily papersupdated Oct 7, 12:10 UTC · 1d agofirst Oct 6, 20:00 UTC · 2d agoAI safety & security 2 sources1

ColNanoVDR: Document-Free Query Distillation for Multi-Vector Visual Document Retrieval via Optimal Transport

ColNanoVDR distills multi-vector document retrievers into 149M students that keep 95% NDCG@5 and run up to 26x faster.

ColNanoVDR distills multi-vector visual document retrievers into 149M text-only students without encoding training pages. Its OTW objective aligns student and teacher query tokens using entropic optimal transport and learned token weights, and the alignment cost is proven to bound MaxSim differences on every page. Students distilled from five state-of-the-art teachers retain about 95% of teacher NDCG@5 on ViDoRe v1-v3 while encoding queries up to 26x faster. Under identical training, OTW matches score distillation while reading 12.6x less cached teacher data.

Hugging Face daily papersSep 27, 20:00 UTC · 11d agoAI research

Generative Late-Interaction Embeddings For Visual Document Retrieval

GLIE compresses visual document retrieval embeddings to four vectors per page while retaining nearly 80% of uncompressed nDCG@5 accuracy.

Researchers analyzing late-interaction retrieval embeddings found they lie exactly on the unit sphere and concentrate near a manifold of intrinsic dimension five to six. GLIE learns a few k vectors per page that serve as a lightweight index and a basis to regenerate the full embedding set for exact rescoring of top candidates at query time. On ViDoRe v1 with four vectors per page, GLIE retains nearly 80% of uncompressed nDCG@5 versus 70% for the best prior post-hoc method, using a 415K-parameter network trained in under three GPU-minutes on 1,000 pages.

Hugging Face daily papersSep 9, 20:00 UTC · 29d agoAI research1

ViDoRe

Timeline4 articles

Appears with

Inverting Multi-Vector Visual Document Indices

Inverting Multi-Vector Visual Document Indices

ColNanoVDR: Document-Free Query Distillation for Multi-Vector Visual Document Retrieval via Optimal Transport

Generative Late-Interaction Embeddings For Visual Document Retrieval