mlsanityData to 14 Oct 2026

G2-Reader: Dual Evolving Graphs for Multimodal Document QA

arXiv 2601.22055 ↗First posted 29 Jan 2026cs.CLcited by 1

On the map: Retrieval-augmented generation → Multimodal RAG

Abstract (arXiv)

Retrieval-augmented generation is a practical paradigm for question answering over long documents, but it remains brittle for multimodal reading where text, tables, and figures are interleaved across many pages. First, flat chunking breaks document-native structure and cross-modal alignment, yielding semantic fragments that are hard to interpret in isolation. Second, even iterative retrieval can fail in long contexts by looping on partial evidence or drifting into irrelevant sections as noise accumulates, since each step is guided only by the current snippet without a persistent global search state. We introduce G2-Reader, a dual-graph system, to address both issues. It evolves a Content Graph to preserve document-native structure and cross-modal semantics, and maintains a Planning Graph, an agentic directed acyclic graph of sub-questions, to track intermediate findings and guide stepwise navigation for evidence completion. On VisDoMBench across five multimodal domains, G2-Reader with Qwen3-VL-32B-Instruct reaches 66.21% average accuracy, outperforming strong baselines and a standalone GPT-5 (53.08%).

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Multimodal RAG, one of 15 directions in Retrieval-augmented generation. Builds on 42 of its 47 references that are in the map. First posted on arXiv 29 Jan 2026.

Builds on

42 of 47 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025RAG-Anything: All-in-One RAG Framework14 Oct
  2. 2025ViDoRAG: Visual Document Retrieval-Augmented Generation via Dynamic Iterative Reasoning Agents25 Feb
  3. 2024VisDoM: Multi-Document QA with Visually Rich Elements Using Multimodal Retrieval-Augmented Generation14 Dec
  4. 2024M3DocRAG: Multi-modal Retrieval is What You Need for Multi-page Multi-document Understanding7 Nov
  5. 2024VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documents14 Oct

Other references · 37

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  3. 2020Longformer: The Long-Document Transformercited by 3,413
  4. 2023A Survey of Large Language Modelscited by 2,576
  5. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,141
  6. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,839
  7. 2024From Local to Global: A Graph RAG Approach to Query-Focused Summarizationcited by 1,453
  8. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,210
  9. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,075
  10. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,063
  11. 2022Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questionscited by 923
  12. 2025Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memorycited by 779
25 more
  1. 2023Active Retrieval Augmented Generationcited by 516
  2. 2021A Dataset of Information-Seeking Questions and Answers Anchored in Research Paperscited by 486
  3. 2023Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergycited by 346
  4. 2024LightRAG: Simple and Fast Retrieval-Augmented Generationcited by 315
  5. 2023Effective Long-Context Scaling of Foundation Modelscited by 303
  6. 2024MinerU: An Open-Source Solution for Precise Document Content Extractioncited by 267
  7. 2020A Note on Over-Smoothing for Graph Neural Networkscited by 249
  8. 2025DeepSeek-OCR: Contexts Optical Compressioncited by 183
  9. 2023SlideVQA: A Dataset for Document Visual Question Answering on Multiple Imagescited by 179
  10. 2021FeTaQA: Free-form Table Question Answeringcited by 175
  11. 2024SPIQA: A Dataset for Multimodal Question Answering on Scientific Paperscited by 75
  12. 2023SciGraphQA: A Large-Scale Synthetic Multi-Turn Question-Answering Dataset for Scientific Graphscited by 68
  13. 2024Auto-RAG: Autonomous Retrieval-Augmented Generation for Large Language Modelscited by 43
  14. 2024UDA: A Benchmark Suite for Retrieval Augmented Generation in Real-world Document Analysiscited by 40
  15. 2024GeAR: Graph-enhanced Agent for Retrieval-augmented Generationcited by 17
  16. 2025MMGraphRAG: Bridging Vision and Language with Interpretable Multimodal Knowledge Graphscited by 16
  17. 2024Graph Neural Network Enhanced Retrieval for Question Answering of LLMscited by 10
  18. 2025Cog-RAG: Cognitive-Inspired Dual-Hypergraph with Theme Alignment Retrieval-Augmented Generationcited by 9
  19. 2025InfoAgent: Advancing Autonomous Information-Seeking Agentscited by 7
  20. 2025Beyond Chunks and Graphs: Retrieval-Augmented Generation through Triplet-Driven Thinkingcited by 6
  21. 2025Interact-RAG: Reason and Interact with the Corpus, Beyond Black-Box Retrievalcited by 6
  22. 2025Scaling External Knowledge Input Beyond Context Windows of LLMs via Multi-Agent Collaborationcited by 4
  23. 2025EcphoryRAG: Re-Imagining Knowledge-Graph RAG via Human Associative Memorycited by 4
  24. 2017Unsupervised Adaptive Re-identification in Open World Dynamic Camera Networkscited by 2
  25. 2025Fishing for Answers: Exploring One-shot vs. Iterative Retrieval Strategies for Retrieval Augmented Generationcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026MAGE-RAG: Multigranular Adaptive Graph Evidence for Agentic Multimodal RAG in Long-Document QAcited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2026HVM-GraphRAG: Holistic-View Multimodal Graph Retrieval-Augmented Generation on Complex Document
  2. 2026MLDocRAG: Multimodal Long-Context Document Retrieval Augmented Generation
  3. 2026DocNavRAG: Document-Structured Graph RAG with Stateful Evidence Construction for Complex Document Question Answering
  4. 2026FLOWREADER: Min-Cost Flow Optimization for Multi-Modal Long Document Q&A
  5. 2026Locate, Judge, Read: Role-Specialized Tool Agents over a Knowledge-Point Index for Open-Domain Multimodal Document QA
  6. 2026TopoGraphRAG-Bench: Evaluating Multimodal GraphRAG on Layout-Grounded Evidence Reasoning
  7. 2026MARA: A Multimodal Adaptive Retrieval-Augmented Framework for Document Question Answering
  8. 2026DocTrace: Towards Traceable Long Document VQA via Hierarchical Evidence Graph Reasoning