mlsanityData to 14 Oct 2026

TAP-RAG: Task-Aware Policy Control for Long-Document Multimodal Question Answering

arXiv 2607.18917 ↗First posted 21 Jul 2026cs.CVcited by 0

On the map: Retrieval-augmented generation → Multimodal RAG

Abstract (arXiv)

Long-document multimodal question answering requires more than retrieving relevant chunks from a large document. Different queries require different evidence behavior. Existing multimodal RAG systems improve evidence access through text chunks, page images, graph links, or heterogeneous document elements, but they often apply a largely query-agnostic evidence-use strategy. We present TAP-RAG, a task-aware policy-controlled RAG framework for long-document multimodal QA. TAP-RAG contains a main controller, the Task-Aware Policy Controller (TAPC), and two policy-guided evidence executors: Task-Aware Query-Guided Flow Diffusion (TA-QFD) and Task-Aware Visual Enhancement (TAVE). For each query, TAPC predicts the task prior, estimates visual/local/global evidence signals, and produces an executable policy. TA-QFD then expands textual and structural evidence over the multimodal document graph, while TAVE selectively inspects page images when visual or layout evidence is needed. A guarded synthesis stage fuses text, visual, and structural evidence and abstains when support is insufficient. On DocBench and MMLongBench-Doc, TAP-RAG achieves the best overall accuracy among the compared systems, improving over a matched multimodal-RAG baseline by +9.1 points (61.1 to 70.2) and +4.5 points (42.2 to 46.7), respectively.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Multimodal RAG, one of 15 directions in Retrieval-augmented generation. Builds on 9 of its 19 references that are in the map. First posted on arXiv 21 Jul 2026.

Builds on

9 of 19 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025RAG-Anything: All-in-One RAG Framework14 Oct
  2. 2024M3DocRAG: Multi-modal Retrieval is What You Need for Multi-page Multi-document Understanding7 Nov
  3. 2024VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documents14 Oct

Other references · 6

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  2. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,839
  3. 2024From Local to Global: A Graph RAG Approach to Query-Focused Summarizationcited by 1,453
  4. 2024Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexitycited by 451
  5. 2024MMLongBench-Doc: Benchmarking Long-context Document Understanding with Visualizationscited by 183
  6. 2025MMGraphRAG: Bridging Vision and Language with Interpretable Multimodal Knowledge Graphscited by 16

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Hierarchical Evidence-Driven Reasoning for Long Document Understanding
  2. 2026DocTrace: Towards Traceable Long Document VQA via Hierarchical Evidence Graph Reasoning
  3. 2026VLD-RAG: Agentic Vision-Language Retrieval-Augmented Generation for Long, Visually-Rich Multi-Page Documents
  4. 2026HierDoc: Hierarchical Page-to-Region Evidence Routing for Long-Document Visual Question Answering
  5. 2026Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering
  6. 2026AutothinkRAG: Complexity-Aware Control of Retrieval-Augmented Reasoning for Image-Text Interaction
  7. 2025Benchmarking Retrieval-Augmented Multimodal Generation for Document Question Answering
  8. 2026TopoGraphRAG-Bench: Evaluating Multimodal GraphRAG on Layout-Grounded Evidence Reasoning