mlsanityData to 14 Oct 2026

From PDF to Evidence: Structure-Aware Retrieval for Clinical Practice Guidelines

arXiv 2609.33447 ↗First posted 27 Sep 2026cs.IR cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Guideline documents are published as unstructured PDFs whose evidence is locked in visual structures---tables, flowcharts, and graded recommendations---that standard retrieval pipelines flatten into fixed-size text chunks. We cast evidence access as a document image analysis problem: parse each page image into typed structural elements, then retrieve structure-aware evidence units that follow the document's own layout (sections, table rows, flowchart paths, graded recommendations), each keeping its structural context so a result points to a specific element rather than a page. On 26 clinical practice guidelines from 9 sources (3,619 pages, Chinese and English) with 199 evidence queries, structure-aware units rank the gold element first under BM25, dense, and hybrid retrieval (hybrid Element Hit@1 of 0.382), with a significant element-level ranking gain over per-element OCR text (MRR_e +0.107, p=0.002; the Hit@5 gain is directional, p=0.17), while matching page-level recall (Page Hit@5 0.879 vs. 0.889, p=0.75) at 3.8x less context and clearly outperforming a ColPali visual-RAG baseline (PH@5 0.497).

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 23 of its 29 references that are in the map. First posted on arXiv 27 Sep 2026.

Builds on

23 of 29 references.

References · 23

  1. 2025Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Modelscited by 1,261
  2. 2019PubLayNet: largest dataset ever for document layout analysiscited by 351
  3. 2024Benchmarking Retrieval-Augmented Generation for Medicinecited by 331
  4. 2024ColPali: Efficient Document Retrieval with Vision Language Modelscited by 195
  5. 2025DeepSeek-OCR: Contexts Optical Compressioncited by 183
  6. 2024OmniDocBench: Benchmarking Diverse PDF Document Parsing with Comprehensive Annotationscited by 150
  7. 2022DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysiscited by 148
  8. 2023Almanac: Retrieval-Augmented Language Models for Clinical Medicinecited by 123
  9. 2022TableFormer: Table Structure Understanding with Transformerscited by 94
  10. 2025dots.ocr: Multilingual Document Layout Parsing in a Single Vision-Language Modelcited by 47
  11. 2024Improving Medical Reasoning through Retrieval and Self-Reflection with Retrieval-Augmented Large Language Modelscited by 34
  12. 2023MedDM:LLM-executable clinical guidance tree for clinical decision-makingcited by 19
11 more
  1. 2024Detect-Order-Construct: A Tree Construction based Approach for Hierarchical Document Structure Analysiscited by 15
  2. 2025RegionRAG: Region-level Retrieval-Augmented Generation for Visual Document Understandingcited by 8
  3. 2026Unlimited OCR Workscited by 8
  4. 2020GGPONC: A Corpus of German Medical Text with Rich Metadata Based on Clinical Practice Guidelinescited by 7
  5. 2025Grounding Large Language Models in Clinical Evidence: A Retrieval-Augmented Generation System for Querying UK NICE Clinical Guidelinescited by 7
  6. 2026A Decade-Scale Benchmark Evaluating LLMs' Clinical Practice Guidelines Detection and Adherence in Multi-turn Conversationscited by 3
  7. 2026Guideline2Graph: Profile-Aware Multimodal Parsing for Executable Clinical Decision Graphscited by 2
  8. 2026ClinicBot: A Guideline-Grounded Clinical Chatbot with Prioritized Evidence RAG and Verifiable Citationscited by 2
  9. 2026From PDF to RAG-Ready: Evaluating Document Conversion Frameworks for Domain-Specific Question Answeringcited by 1
  10. 2026SPIRE: Structure-Preserving Interpretable Retrieval of Evidencecited by 1
  11. 2026SEM-RAG: Structure-Preserving Multimodal Graph Compilation and Entropy-Guided Retrieval for Telecommunication Standardscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Graph-Aware Late Chunking for Retrieval-Augmented Generation in Biomedical Literature
  2. 2026DocMemo: Dynamic Evidence Discovery via Probabilistic Memory-Guided Retrieval for Multi-Modal Document Understanding
  3. 2026TopoGraphRAG-Bench: Evaluating Multimodal GraphRAG on Layout-Grounded Evidence Reasoning
  4. 2026LITTA: Late-Interaction and Test-Time Alignment for Visually-Grounded Multimodal Retrieval
  5. 2026Found but Not Read: When Extracted Text Closes the Retrieval-Reading Gap in Document Vision-Language Models
  6. 2025Query-driven Document-level Scientific Evidence Extraction from Biomedical Studies
  7. 2026VELD-QA: A Large-Scale Dataset for Visual Evidence Grounding and Question Answering in Long Documents
  8. 2025MedRepBench: A Comprehensive Benchmark for Medical Report Interpretation