mlsanityData to 7 Oct 2026

DocPruner: A Storage-Efficient Framework for Multi-Vector Visual Document Retrieval via Adaptive Patch-Level Embedding Pruning

arXiv 2509.23883 ↗First posted 28 Sep 2025cs.CL cs.IRcited by 19

Not on the map. Why?

Abstract (arXiv)

Visual Document Retrieval (VDR), the task of retrieving visually-rich document pages using queries that combine visual and textual cues, is crucial for numerous real-world applications. Recent state-of-the-art methods leverage Large Vision-Language Models (LVLMs) in a multi-vector paradigm, representing each document as patch-level embeddings to capture fine-grained details. While highly effective, this approach introduces a critical challenge: prohibitive storage overhead, as storing hundreds of vectors per page makes large-scale deployment costly and impractical. To address this, we introduce DocPruner, the first framework to employ adaptive patch-level embedding pruning for VDR to effectively reduce the storage overhead. DocPruner leverages the intra-document patch attention distribution to dynamically identify and discard redundant embeddings for each document. This adaptive mechanism enables a significant 50-60% reduction in storage for leading multi-vector VDR models with negligible degradation in document retrieval performance. Extensive experiments across more than ten representative datasets validate that DocPruner offers a robust, flexible, and effective solution for building storage-efficient, large-scale VDR systems.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). Builds on 96 of its 110 references that are in the map. First posted on arXiv 28 Sep 2025.

Builds on

96 of 110 references.

References · 96

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,069
  3. 2025Qwen2.5-VL Technical Reportcited by 5,418
  4. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  5. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,669
  6. 2020ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERTcited by 1,160
  7. 2024M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillationcited by 1,050
  8. 2022Token Merging: Your ViT But Fastercited by 995
  9. 2015Deep Learning and the Information Bottleneck Principlecited by 793
  10. 2024PaliGemma: A versatile 3B VLM for transfercited by 679
  11. 2024An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Modelscited by 629
  12. 2024LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Modelscited by 571
38 more
  1. 2023C-Pack: Packed Resources For General Chinese Embeddingscited by 556
  2. 2021ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interactioncited by 464
  3. 2024Hallucination of Multimodal Large Language Models: A Surveycited by 380
  4. 2024SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inferencecited by 378
  5. 2023Improving Text Embeddings with Large Language Modelscited by 295
  6. 2024VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Taskscited by 222
  7. 2024mPLUG-DocOwl 1.5: Unified Structure Learning for OCR-free Document Understandingcited by 220
  8. 2023A Survey on Deep Neural Network Pruning-Taxonomy, Comparison, Analysis, and Recommendationscited by 200
  9. 2021Whitening Sentence Representations for Better Semantics and Faster Retrievalcited by 198
  10. 2025Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontierscited by 193
  11. 2024ColPali: Efficient Document Retrieval with Vision Language Modelscited by 190
  12. 2024Agent AI: Surveying the Horizons of Multimodal Interactioncited by 167
  13. 2024GME: Improving Universal Multimodal Retrieval by Multimodal LLMscited by 166
  14. 2024MM-Embed: Universal Multimodal Retrieval with Multimodal LLMscited by 159
  15. 2024Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMscited by 153
  16. 2024Exploring the Reasoning Abilities of Multimodal Large Language Models (MLLMs): A Comprehensive Survey on Emerging Trends in Multimodal Reasoningcited by 136
  17. 2024The Revolution of Multimodal Large Language Models: A Surveycited by 134
  18. 2024Fit and Prune: Fast and Training-free Visual Token Pruning for Multi-modal Large Language Modelscited by 125
  19. 2024xRAG: Extreme Context Compression for Retrieval-augmented Generation with One Tokencited by 123
  20. 2024mPLUG-DocOwl2: High-resolution Compressing for OCR-free Multi-page Document Understandingcited by 113
  21. 2025VLM2Vec-V2: Advancing Multimodal Embedding for Videos, Images, and Visual Documentscited by 110
  22. 2024Unifying Multimodal Retrieval via Document Screenshot Embeddingcited by 107
  23. 2024Large Multimodal Agents: A Surveycited by 99
  24. 2022PLAID: An Efficient Engine for Late Interaction Retrievalcited by 95
  25. 2024ATP-LLaVA: Adaptive Token Pruning for Large Vision Language Modelscited by 89
  26. 2024RAP: Retrieval-Augmented Planning with Contextual Memory for Multimodal LLM Agentscited by 80
  27. 2023UrbanCLIP: Learning Text-enhanced Urban Region Profiling with Contrastive Language-Image Pretraining from the Webcited by 77
  28. 2025Token Pruning in Multimodal Large Language Models: Are We Solving the Right Problem?cited by 76
  29. 2025jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrievalcited by 72
  30. 2025ViDoRe Benchmark V2: Raising the Bar for Visual Retrievalcited by 58
  31. 2025VDocRAG: Retrieval-Augmented Generation over Visually-Rich Documentscited by 57
  32. 2024A Survey of Mathematical Reasoning in the Era of Multimodal Large Language Model: Benchmark, Method & Challengescited by 55
  33. 2025MMDocIR: Benchmarking Multimodal Retrieval for Long Documentscited by 43
  34. 2024Mitigating Modality Prior-Induced Hallucinations in Multimodal Large Language Models via Deciphering Attention Causalitycited by 42
  35. 2025Ask in Any Modality: A Comprehensive Survey on Multimodal Retrieval-Augmented Generationcited by 42
  36. 2023Cross-Modal Retrieval: A Systematic Review of Methods and Future Directionscited by 40
  37. 2024ErrorRadar: Benchmarking Complex Mathematical Reasoning of Multimodal Large Language Models Via Error Detectioncited by 37
  38. 2025A Survey of Multimodal Retrieval-Augmented Generationcited by 37

50 most cited of 96 All 96 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 15.

19 papers.

Sort:

By area:

  1. 2025Don't Just Chase "Highlighted Tokens" in MLLMs: Revisiting Visual Holistic Context Retentioncited by 45
  2. 2025Scaling Beyond Context: A Survey of Multimodal Retrieval-Augmented Generation for Document Understandingcited by 9
  3. 2026Sculpting the Vector Space: Towards Efficient Multi-Vector Visual Document Retrieval via Prune-then-Merge Frameworkcited by 8
  4. 2026Unlocking Multimodal Document Intelligence: From Current Triumphs to Future Frontiers of Visual Document Retrievalcited by 8
  5. 2026Structural Anchor Pruning: Training-Free Multi-Vector Compression for Visual Document Retrievalcited by 6
  6. 2026CausalEmbed: Auto-Regressive Multi-Vector Generation in Latent Space for Visual Document Embeddingcited by 6
  7. 2026Visual Late Chunking: An Empirical Study of Contextual Chunking for Efficient Visual Document Retrievalcited by 6
  8. 2025Hybrid-Vector Retrieval for Visually Rich Documents: Combining Single-Vector Efficiency and Multi-Vector Accuracycited by 4
  9. 2026Beyond the Grid: Layout-Informed Multi-Vector Retrieval with Parsed Visual Document Representationscited by 3
  10. 2025Roles of MLLMs in Visually Rich Document Retrieval for RAG: A Surveycited by 2
  11. 2026NanoVDR: Distilling a 2B Vision-Language Retriever into a 70M Text-Only Encoder for Visual Document Retrievalcited by 2
  12. 2026MM-Matryoshka: Towards Budget-Elastic Visual Document Retrieval via a 2D Multimodal Matryoshka Training Frameworkcited by 2
7 more
  1. 2026LEMUR: Learned Multi-Vector Retrievalcited by 1
  2. 2026AnchorFold: A Focus-Then-Fold Framework via Recursive Attention Propagation for Efficient Multi-Vector Visual Document Retrievalcited by 1
  3. 2026Spatial Matryoshka Training for Multi-Granularity Visual Document Retrievalcited by 0
  4. 2026Doc-REFRAG: Rethinking Multimodal Document Retrieval-Augmented Generationcited by 0
  5. 2026EigenLI: Spectral Approximations to Late Interactioncited by 0
  6. 2026MAGIC: Marginal-Guided Compression with Optimal Transport for Efficient Visual Document Retrievalcited by 0
  7. 2026AdaMerge: Tuning-Free Patch Compression for Multi-Vector Visual Document Retrievalcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026ColNanoVDR: Document-Free Query Distillation for Multi-Vector Visual Document Retrieval via Optimal Transport
  2. 2026MURE: Hierarchical Multi-Resolution Encoding via Vision-Language Models for Visual Document Retrieval
  3. 2025Reproducibility, Replicability, and Insights into Visual Document Retrieval with Late Interaction
  4. 2026ReAlign: Optimizing the Visual Document Retriever with Reasoning-Guided Fine-Grained Alignment
  5. 2026LightSTAR: Efficient Visual Document Retrieval via Lightweight Selection with Vision-Adaptive Refinement
  6. 2026Inference-Free Multimodal Learned Sparse Retrieval for Production-Scale Visual Document Search
  7. 2025SERVAL: Surprisingly Effective Zero-Shot Visual Document Retrieval Powered by Large Vision and Language Models
  8. 2025ModernVBERT: Towards Smaller Visual Document Retrievers