Unlocking Multimodal Document Intelligence: From Current Triumphs to Future Frontiers of Visual Document Retrieval
cs.CL, cs.IR
Submitted: 2026-02-23
Updated: 2026-09-07
Comments: Under review. This version updates the relevant works released before 25 August, 2026
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- MM-BRIGHT: A Multi-Task Multimodal Benchmark for Reasoning-Intensive Retrieval
- Agent-X: Evaluating Deep Multimodal Reasoning in Vision-Centric Agentic Tasks
- Qwen2.5-VL Technical Report
- PaliGemma: A versatile 3B VLM for transfer
- mmE5: Improving Multimodal Multilingual Embeddings via High-quality Synthetic Data
- VisR-Bench: An Empirical Study on Visual Retrieval-Augmented Generation for Multilingual Long Document Understanding
- Why Reasoning Matters? A Survey of Advancements in Multimodal Reasoning (v1)
- ReasonEmbed: Enhanced Text Embeddings for Reasoning-Intensive Document Retrieval
- Learning Multi-Level Features with Matryoshka Sparse Autoencoders
- MLLM Is a Strong Reranker: Advancing Multimodal Retrieval-augmented Generation via Knowledge-enhanced Reranking and Noise-injected Training
- When Vision Meets Texts in Listwise Reranking
- BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent
- Matryoshka Multimodal Models
- Recent advances in text embedding: A Comprehensive Review of Top-Performing Methods on the MTEB Benchmark
- A Survey on Knowledge-Oriented Retrieval-Augmented Generation
- Toward Generalizable Evaluation in the LLM Era: A Survey Beyond Benchmarks
- M3DocRAG: Multi-modal Retrieval is What You Need for Multi-page Multi-document Understanding
- ReinPool: Reinforcement Learning Pooling Multi-Vector Embeddings for Retrieval System
- Maintaining MTEB: Towards Long Term Usability and Reproducibility of Embedding Benchmarks
- Document AI: Benchmarks, Models and Applications
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering