# Visual Document Understanding and Reasoning: A Multi-Agent Collaboration Framework with Agent-Wise Adaptive Test-Time Scaling

arXiv: https://arxiv.org/abs/2508.03404 · HTML: https://mlsanity.com/paper/2508.03404/

## Key facts

- Cited by 30 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 14 of its 37 references that are in the map.
- First posted on arXiv 5 Aug 2025.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Multimodal LLMs / vision-language models / [Document understanding and multimodal RAG](https://mlsanity.com/mllm/document-understanding-and-multimodal-rag/)
- LLM agents and tool use / [Multimodal and embodied agents](https://mlsanity.com/agents/multimodal-and-embodied-agents/)

## Builds on (most cited first, 14 of 14)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [s1: Simple test-time scaling](https://mlsanity.com/paper/2501.19393/) (2025)
- [LMMs-Eval: Reality Check on the Evaluation of Large Multimodal Models](https://mlsanity.com/paper/2407.12772/) (2024)
- [A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?](https://mlsanity.com/paper/2503.24235/) (2025)
- [VisualPRM: An Effective Process Reward Model for Multimodal Reasoning](https://mlsanity.com/paper/2503.10291/) (2025)
- [MiMo: Unlocking the Reasoning Potential of Language Model -- From Pretraining to Posttraining](https://mlsanity.com/paper/2505.07608/) (2025)
- [MiMo-VL Technical Report](https://mlsanity.com/paper/2506.03569/) (2025)
- [Scaling Test-Time Compute Without Verification or RL is Suboptimal](https://mlsanity.com/paper/2502.12118/) (2025)
- [TableVQA-Bench: A Visual Question Answering Benchmark on Multiple Table Domains](https://mlsanity.com/paper/2404.19205/) (2024)
- [METAL: A Multi-Agent Framework for Chart Generation with Test-Time Scaling](https://mlsanity.com/paper/2502.17651/) (2025)
- [Skywork-VL Reward: An Effective Reward Model for Multimodal Understanding and Reasoning](https://mlsanity.com/paper/2505.07263/) (2025)

## Built on by (most cited first, 30 of 30)

- [ReTrack: Evidence-Driven Dual-Stream Directional Anchor Calibration Network for Composed Video Retrieval](https://mlsanity.com/paper/2604.17898/) (2026)
- [A Survey of Data Agents: Emerging Paradigm or Overstated Hype?](https://mlsanity.com/paper/2510.23587/) (2025)
- [Explain Before You Answer: A Survey on Compositional Visual Reasoning](https://mlsanity.com/paper/2508.17298/) (2025)
- [INTENT: Invariance and Discrimination-aware Noise Mitigation for Robust Composed Image Retrieval](https://mlsanity.com/paper/2604.18051/) (2026)
- [HABIT: Chrono-Synergia Robust Progressive Learning Framework for Composed Image Retrieval](https://mlsanity.com/paper/2604.18037/) (2026)
- [DocLens : A Tool-Augmented Multi-Agent Framework for Long Visual Document Understanding](https://mlsanity.com/paper/2511.11552/) (2025)
- [DocDancer: Towards Agentic Document-Grounded Information Seeking](https://mlsanity.com/paper/2601.05163/) (2026)
- [Dual Latent Memory for Visual Multi-agent System](https://mlsanity.com/paper/2602.00471/) (2026)
- [SkillGraph: Self-Evolving Multi-Agent Collaboration with Multimodal Graph Topology](https://mlsanity.com/paper/2604.17503/) (2026)
- [ASTGI: Adaptive Spatio-Temporal Graph Interactions for Irregular Multivariate Time Series Forecasting](https://mlsanity.com/paper/2509.23313/) (2025)
- [Diffusion-Based Image Editing for Breaking Robust Watermarks](https://mlsanity.com/paper/2510.05978/) (2025)
- [PATRA: Pattern-Aware Alignment and Balanced Reasoning for Time Series Question Answering](https://mlsanity.com/paper/2602.23161/) (2026)
- [From Experience to Skill: Multi-Agent Generative Engine Optimization via Reusable Strategy Learning](https://mlsanity.com/paper/2604.19516/) (2026)
- [Diffusion-Based Image Editing: An Unforeseen Adversary to Robust Invisible Watermarks](https://mlsanity.com/paper/2511.05598/) (2025)
- [Insight-V++: Towards Advanced Long-Chain Visual Reasoning with Multimodal Large Language Models](https://mlsanity.com/paper/2603.18118/) (2026)
- [Hierarchical Evidence-Driven Reasoning for Long Document Understanding](https://mlsanity.com/paper/2607.04625/) (2026)
- [MVCL-DAF++: Enhancing Multimodal Intent Recognition via Prototype-Aware Contrastive Alignment and Coarse-to-Fine Dynamic Attention Fusion](https://mlsanity.com/paper/2509.17446/) (2025)
- [M3Prune: Hierarchical Communication Graph Pruning for Efficient Multi-Modal Multi-Agent Retrieval-Augmented Generation](https://mlsanity.com/paper/2511.19969/) (2025)
- [Decoupling Stability and Plasticity for Multi-Modal Test-Time Adaptation](https://mlsanity.com/paper/2603.00574/) (2026)
- [IMAGINE: Adaptive Schema-Imagery Enhanced Composition for Composed Video Retrieval](https://mlsanity.com/paper/2606.08144/) (2026)
- [SPOT-E: Test-Time Entropy Shaping with Visual Spotlights for Frozen VLMs](https://mlsanity.com/paper/2606.20244/) (2026)
- [Adapting LLMs to Time Series Forecasting via Temporal Heterogeneity Modeling and Representation Alignment](https://mlsanity.com/paper/2508.07195/) (2025)
- [Robust Concept Erasure in Diffusion Models: A Theoretical Perspective on Security and Robustness](https://mlsanity.com/paper/2509.12024/) (2025)
- [DeRainMamba: A Frequency-Aware State Space Model with Detail Enhancement for Image Deraining](https://mlsanity.com/paper/2510.06746/) (2025)
- [Diagnosing Failure Modes of Shared-State Collaboration in Resource-Constrained Visual Agents](https://mlsanity.com/paper/2605.31354/) (2026)
- [DocAtlas: Long-Document Understanding as Mutable-State Interaction](https://mlsanity.com/paper/2608.07527/) (2026)
- [D2-ScaleAgent: Dual-Dimensional Scaling for Long Document Understanding](https://mlsanity.com/paper/2608.16417/) (2026)
- [An Empirical Study of VLM Pipelines for Long-Document QA](https://mlsanity.com/paper/2609.29933/) (2026)
- [KVCMAS: Efficient KV cache Correction for Shared Context in Multi-Agent Systems](https://mlsanity.com/paper/2609.34060/) (2026)
- [It Takes Workflows to Evolve Better Workflows](https://mlsanity.com/paper/2610.01026/) (2026)

## Abstract (arXiv)

The dominant paradigm of monolithic scaling in Vision-Language Models (VLMs) is failing for understanding and reasoning in documents, yielding diminishing returns as it struggles with the inherent need of this domain for document-based procedural reasoning, cognitive complexity, and factual accuracy. To this end, we introduce MACT, a Multi-Agent Collaboration framework with agent-wise adaptive Test-time scaling that pioneers a paradigm shift to procedural scaling, adapting dynamically to the functional entities of visual documents understanding and reasoning. MACT decomposes the visual document processing flow into four specialized agents, i.e., planning, execution, judgment, and answer, to resolve cognitive overload and introduce a critical self-correction loop for factual grounding. This collaborative architecture is amplified by an agent-wise adaptive test-time scaling strategy that intelligently allocates computational resources based on the complexity and redundancy of each functionality. Evaluated on multiple visual document understanding benchmarks, MACT achieves superior performance with a smaller parameter scale, adapting effectively to various document scenarios without compromising its general or mathematical reasoning capabilities. The three variants of MACT consistently attain top-three average performance rankings, with average performance enhancements of 9.9-11.5% over the base models. The source code will be released publicly.