mlsanityData to 7 Oct 2026

InSight-doc: Agentic Visual Perception for Long-Document Understanding

arXiv 2608.10628 ↗First posted 11 Aug 2026cs.CV cs.CL cs.LGcited by 0

On the map: LLM agents and tool use → Multimodal and embodied agents

Abstract (arXiv)

Long-document understanding often requires reasoning over many visually rich pages, making inference costly and prone to context rot. In this work, we propose InSight-doc, an agentic visual perception framework that treats visual resolution as an adaptive reasoning-time resource. InSight-doc starts from low resolution and selectively zooms into high-resolution regions for finer evidence, without relying on any external retriever. To train such an agent, we construct an active-perception corpus of 17.9K high-quality SFT examples with region-level zoom-in trajectories, accompanied by 19.2K hard RL examples. Through SFT+RL, InSight-doc-8B improves the baseline by 4.3--16.4 accuracy points over document VQA benchmarks. On long documents, it reduces hallucination by more than 40% and inference latency by 41%--68% while maintaining an accuracy lead. Our code, datasets, and model are released at https://github.com/m-Just/InSight-doc .

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Multimodal and embodied agents, one of 16 directions in LLM agents and tool use. Builds on 33 of its 47 references that are in the map. First posted on arXiv 11 Aug 2026.

Builds on

33 of 47 references.

References · 33

  1. 2017Attention Is All You Needcited by 59,932
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  3. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  4. 2022Large Language Models are Zero-Shot Reasonerscited by 4,323
  5. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 1,980
  6. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,419
  7. 2020DocVQA: A Dataset for VQA on Document Imagescited by 1,288
  8. 2023V*: Guided Visual Search as a Core Mechanism in Multimodal LLMscited by 459
  9. 2025Pixel Reasoner: Incentivizing Pixel-Space Reasoning with Curiosity-Driven Reinforcement Learningcited by 267
  10. 2024MinerU: An Open-Source Solution for Precise Document Content Extractioncited by 255
  11. 2024VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documentscited by 237
  12. 2024MMLongBench-Doc: Benchmarking Long-context Document Understanding with Visualizationscited by 174
21 more
  1. 2023Document Understanding Dataset and Evaluation (DUDE)cited by 132
  2. 2025GRIT: Teaching MLLMs to Think with Imagescited by 121
  3. 2025Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Searchcited by 116
  4. 2024mPLUG-DocOwl2: High-resolution Compressing for OCR-free Multi-page Document Understandingcited by 113
  5. 2024M3DocRAG: Multi-modal Retrieval is What You Need for Multi-page Multi-document Understandingcited by 110
  6. 2024Unifying Multimodal Retrieval via Document Screenshot Embeddingcited by 107
  7. 2024ZoomEye: Enhancing Multimodal LLMs with Human-Like Zooming Capabilities through Tree-Based Image Explorationcited by 99
  8. 2024LongDocURL: a Comprehensive Multimodal Long Document Benchmark Integrating Understanding, Reasoning, and Locatingcited by 70
  9. 2025VRAG-RL: Empower Vision-Perception-Based RAG for Visually Rich Information Understanding via Iterative Reasoning with Reinforcement Learningcited by 68
  10. 2025Traceable Evidence Enhanced Visual Grounded Reasoning: Evaluation and Methodologycited by 67
  11. 2025VDocRAG: Retrieval-Augmented Generation over Visually-Rich Documentscited by 57
  12. 2025Paper2Poster: Towards Multimodal Poster Automation from Scientific Paperscited by 56
  13. 2025DyFo: A Training-Free Dynamic Focus Visual Search for Enhancing LMMs in Fine-Grained Visual Understandingcited by 47
  14. 2024SV-RAG: LoRA-Contextualizing Adaptation of MLLMs for Long Document Understandingcited by 25
  15. 2025Simple o3: Towards Interleaved Vision-Language Reasoningcited by 24
  16. 2025MoLoRAG: Bootstrapping Document Understanding via Multi-modal Logic-aware Retrievalcited by 17
  17. 2025DocR1: Evidence Page-Guided GRPO for Multi-Page Document Understandingcited by 11
  18. 2025InSight-o3: Empowering Multimodal Foundation Models with Generalized Visual Searchcited by 9
  19. 2025URaG: Unified Retrieval and Generation in Multimodal LLMs for Efficient Long Document Understandingcited by 5
  20. 2026MM-Doc-R1: Training Agents for Long Document Visual Question Answering through Multi-turn Reinforcement Learningcited by 4
  21. 2026Progressive Video Condensation with MLLM Agent for Long-form Video Understandingcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Visual Document Understanding and Reasoning: A Multi-Agent Collaboration Framework with Agent-Wise Adaptive Test-Time Scaling
  2. 2025Resolving Evidence Sparsity: Agentic Context Engineering for Long-Document Understanding
  3. 2025DocLens : A Tool-Augmented Multi-Agent Framework for Long Visual Document Understanding
  4. 2025Active Video Perception: Iterative Evidence Seeking for Agentic Long Video Understanding
  5. 2026ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both
  6. 2026SynthDocBench: Controlled Benchmark for Long-Context Visual Document Understanding
  7. 2026Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval
  8. 2026DocAtlas: Long-Document Understanding as Mutable-State Interaction