arXiv:2601.20107v3 Announce Type: replace-cross Abstract: Recent Vision-Language Models (e.g., ColPali) enable fine-grained Visual Document Retrieval (VDR) but incur prohibitive multi-vector index st...
Multi-vector vision-language retrievers enable fine-grained Visual Document Retrieval (VDR) through late interaction, but storing and scoring hundreds of visual patch embeddings per page incurs substantial overhead. Existing training-free methods rely on pruning or merging: pruning degrades sharply under aggressive compression, whereas merging does not explicitly prioritize important regions when forming representatives.
arXiv:2609.21018v1 Announce Type: new Abstract: Recent visual document retrieval (VDR) systems such as ColPali use multi-vector page embeddings, in which patch-level vectors enable fine-grained evide...
arXiv:2606. 13141v1 Announce Type: new Abstract: Retrieval-augmented generation is moving beyond text into long, egocentric video, where systems must select query-relevant chunks across multiple modalities and temporal granularities.
arXiv:2609.15320v1 Announce Type: new Abstract: Volume-based multimodal retrieval jointly scores a text query with a candidate's video, audio, and subtitle embeddings. While this approach captures hi...
arXiv:2608.29951v1 Announce Type: new Abstract: Multi-modal late-interaction retrievers achieve strong retrieval on visually rich documents by representing each page as per patch embeddings and match...
FastE is a training‑free, plug‑and‑play method that compresses token prefixes in large language model (LLM) embedding inference. It uses a shared fixed threshold on batch‑mean readout‑prefix alignment to decide when to compress and ranks prefix states by readout attention scores to keep the most important ones. Experiments on Qwen3‑Embedding models show that FastE can cut decoder‑backbone FLOPs by over 40% while preserving more than 99% of the original ranking quality across multiple benchmarks and tasks.
arXiv:2607. 16213v1 Announce Type: new Abstract: Large Language Models (LLMs) generate text autoregressively, relying on a key-value (KV) cache whose memory footprint grows linearly with context length, creating a major bottleneck.
arXiv:2505. 17810v2 Announce Type: replace Abstract: Approximate nearest neighbor (ANN) search is a performance-critical component of many machine learning pipelines, and rigorous benchmarking is essential for assessing the performance of vector indexes for ANN search.
arXiv:2609.07262v1 Announce Type: cross Abstract: Late-interaction visual document retrievers preserve fine-grained page evidence by storing many token embeddings per page, but the resulting storage...
arXiv:2608. 16628v1 Announce Type: new Abstract: Modern Multimodal Retrieval-Augmented Generation (M-RAG) systems are fundamentally limited by the binary connectivity paradigm of traditional simple graphs, which fails to capture the intricate, high-order correlations among heterogeneous entities, such as the N-ary relationships between a visual chart, its scattered textual descriptions, and underlying numerical data.
The paper introduces Invoice Haystack, a benchmark of 1,500 anonymized invoices and 200 question‑answer pairs that tests document retrieval and visual question answering under strong visual homogeneity. It shows that existing benchmarks suffer from embedding collapse, with Invoice Haystack’s mean pairwise cosine similarity at 0.73 versus 0.38 and 0.31 in DocHaystack and InfoHaystack. The authors propose VL‑RAG, a hybrid retrieval‑augmented generation framework that combines text and visual embeddings and a VLM‑based verification filter, achieving 60.0% Recall@1 on Invoice Haystack‑500 and improving performance on other benchmarks.