CuraView: A Multi-Agent Framework for Medical Hallucination Detection with GraphRAG-Enhanced Knowledge Verification

• URL: http://arxiv.org/abs/2605.03476v1
• Date: Tue, 05 May 2026 08:05:31 GMT
• Title: CuraView: A Multi-Agent Framework for Medical Hallucination Detection with GraphRAG-Enhanced Knowledge Verification
• Abstract summary: We present CuraView, a framework for sentence-level detection and evidence-grounded explanation of faithfulness hallucinations in discharge summaries.<n>CuraView constructs a GraphRAG-based knowledge graph from patient-level EHRs.<n>We evaluate CuraView on a subset of 250 patients from the Discharge-Me benchmark, with 50 patients held out for testing.
• Score: 3.422186949568493
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Discharge summaries require extracting critical information from lengthy electronic health records (EHRs), a process that is labor-intensive when performed manually. Large language models (LLMs) can improve generation efficiency; however, they are prone to producing faithfulness hallucinations, statements that contradict source records, posing direct risks to patient safety. To address this, we present CuraView, a multi-agent framework for sentence-level detection and evidence-grounded explanation of faithfulness hallucinations in discharge summaries. CuraView constructs a GraphRAG-based knowledge graph from patient-level EHRs and implements a closed-loop generation-detection pipeline with sentence-level evidence retrieval and classification spanning four evidence grades from strong support to direct contradiction (E1-E4), yielding structured and interpretable evidence chains. We evaluate CuraView on a subset of 250 patients from the Discharge-Me benchmark, with 50 patients held out for testing. Our fine-tuned Qwen3-14B detection model achieves an F1 of 0.831 on the safety-critical E4 metric (90.9% recall, 76.5% precision) and an F1 of 0.823 on E3+E4, representing a 50.0% relative improvement over the base model and outperforming RAGTruth-style and QAGS-style baselines. These results demonstrate that evidence-chain-based graph retrieval verification substantially improves the factual reliability of clinical documentation, while simultaneously producing reusable annotated datasets for downstream model training and distillation.

Related papers

• CareGraph: An Auditable Hybrid AI Framework for Evidence-Grounded Personalized Longitudinal Health Intelligence [0.0]
  CareGraph organizes evidence without diagnosing, predicting outcomes, selecting treatment, or making autonomous clinical decisions.<n>Tests used synthetic cohorts of 400 patients each for development, validation, and holdout.<n>CareGraph offers a safety bounded foundation for intelligent personalized health systems.
  arXiv  Detail & Related papers  (2026-08-25T20:03:52Z)
• Faithful by Design: Evaluating and Improving LLM-Generated Clinical Trial Summaries for Multi-Stakeholder Audiences [0.0]
  Large language models are increasingly used to summarize clinical trial results for healthcare providers, patients, and payers, but their tendency to hallucinate poses significant risks in this high-stakes context.<n>This study introduces a benchmark evaluation framework for measuring the faithfulness of LLM-generated clinical trial summaries across three stakeholder audiences.<n> Baseline measurements were established for GPT-4o, Claude Sonnet 4.6, and Gemini 2.5 Flash across 1,800 generated summaries scored using a cross-encoder natural language inference (NLI) model.
  arXiv  Detail & Related papers  (2026-07-10T19:29:26Z)
• Disease-Centric Vision-Language Pretraining with Hybrid Visual Encoding for 3D Computed Tomography [75.7789001619107]
  Vision-language pre-training holds great promise for general-purpose medical AI.<n>We propose a tailored framework featuring three key components.<n>A diagnosis-aware prompt strategy employs real clinical phrases and aggregated disease prototypes to bridge the pre-training-inference gap.
  arXiv  Detail & Related papers  (2026-06-24T08:24:45Z)
• Hallucination Detection and Correction in Medical VLMs via Counter-Evidence Verification [44.48746443825558]
  Vision-Language models (VLMs) reliability in medical diagnosis is challenged by trust-undermining hallucinations.<n>We propose Counter-Evidence Verification (CoEV), a training-free plug-and-play framework that detects and corrects hallucinations.<n>For hallucination detection, CoEV consistently outperforms existing methods, improving average PR-AUC and ROC-AUC.<n>For hallucination correction, it improves Micro-F1 by up to 12.5%, reduces hallucination rates by over 11.9% on medical report generation, and also boosts medical VQA accuracy.
  arXiv  Detail & Related papers  (2026-06-17T02:14:44Z)
• Full-Self Diagnostics (FSD): Physics-Grounded Visual Biomarker Inference from Smartphone Video via Inverse Problems and Operator Learning [0.0]
  Full-Self Diagnostics (FSD) is a unified mathematical framework for recovering latent physiological states from unconstrained 9-second facial videos.<n> Empirical validation on 38812 real-world paired scans across 59 subjects demonstrates practical performance.<n>Results confirm that consumer-grade facial video encodes sufficient structured information for clinically relevant, non-invasive inference under fully unconstrained conditions.
  arXiv  Detail & Related papers  (2026-06-11T20:26:47Z)
• ClinSeekAgent: Automating Multimodal Evidence Seeking for Agentic Clinical Reasoning [49.05361955143476]
  ClinSeekAgent is an automated agentic framework for dynamic multimodal evidence seeking.<n>Given only a clinical query and access to raw data sources, ClinSeekAgent gathers evidence by querying medical knowledge bases, navigating raw EHRs, and invoking medical imaging tools.
  arXiv  Detail & Related papers  (2026-05-19T17:58:37Z)
• ProtoMedAgent: Multimodal Clinical Interpretability via Privacy-Aware Agentic Workflows [7.27267618572654]
  ProtoMedAgent formalizes multimodal clinical reporting as an iterative, zero-gradient test-time optimization problem.<n>Online generation is strictly constrained by exact set-theoretic differentials and a reflective Scribe-Critic loop.<n>ProtoMedAgent achieves 91.2% Comparison Set Faithfulness where it fundamentally outperforms standard RAG (46.2%)
  arXiv  Detail & Related papers  (2026-05-13T20:57:37Z)
• Detecting Clinical Discrepancies in Health Coaching Agents: A Dual-Stream Memory and Reconciliation Architecture [71.46525715889656]
  General-purpose agent memory systems optimize for coherence by overwriting older facts with the user's latest statement.<n>We introduce a Dual-Stream Memory Architecture that strictly separates the patient narrative from the structured clinical record.<n>We evaluate this architecture on 26 patients across 675 longitudinal wellness coaching sessions.
  arXiv  Detail & Related papers  (2026-04-29T17:59:28Z)
• Benchmarking Multi-turn Medical Diagnosis: Hold, Lure, and Self-Correction [72.89352076103889]
  Large language models (LLMs) achieve high accuracy in medical diagnosis when all clinical information is provided in a single turn.<n>We introduce MINT, a high-fidelity, multi-turn medical diagnosis benchmark comprising 1,035 cases.<n>We uncover three persistent behavioral patterns that significantly impact diagnostic decisions.
  arXiv  Detail & Related papers  (2026-04-06T00:23:10Z)
• Causal-Enhanced AI Agents for Medical Research Screening [0.0]
  Systematic reviews are essential for evidence-based medicine, but reviewing 1.5 million+ annual publications manually is infeasible.<n>We present a causal graph-enhanced retrieval-augmented generation system integrating explicit causal reasoning with dual-level knowledge graphs.<n>Our approach enforces evidence-first protocols where every causal claim traces to retrieved literature and automatically generates directed acyclic graphs visualizing intervention-outcome pathways.
  arXiv  Detail & Related papers  (2026-01-06T08:41:16Z)
• Generating Natural-Language Surgical Feedback: From Structured Representation to Domain-Grounded Evaluation [66.7752700084159]
  High-quality feedback from a surgical trainer is pivotal for improving trainee performance and long-term skill acquisition.<n>We present a structure-aware pipeline that learns a surgical action ontology from real trainer-to-trainee transcripts.
  arXiv  Detail & Related papers  (2025-11-19T06:19:34Z)
• EchoBench: Benchmarking Sycophancy in Medical Large Vision-Language Models [82.43729208063468]
  Recent benchmarks for medical Large Vision-Language Models (LVLMs) emphasize leaderboard accuracy, overlooking reliability and safety.<n>We study sycophancy -- models' tendency to uncritically echo user-provided information.<n>We introduce EchoBench, a benchmark to systematically evaluate sycophancy in medical LVLMs.
  arXiv  Detail & Related papers  (2025-09-24T14:09:55Z)
• Evaluating Large Language Models for Evidence-Based Clinical Question Answering [4.101088122511548]
  Large Language Models (LLMs) have demonstrated substantial progress in biomedical and clinical applications.<n>We curate a benchmark drawing from Cochrane systematic reviews and clinical guidelines.<n>We observe consistent performance patterns across sources and clinical domains.
  arXiv  Detail & Related papers  (2025-09-13T15:03:34Z)
• Eyes on the Image: Gaze Supervised Multimodal Learning for Chest X-ray Diagnosis and Report Generation [1.5087814338685968]
  We propose a two-stage framework that enhances disease classification and region-aware radiology report generation from chest X-rays.<n>In the first stage, we introduce a gaze-guided contrastive learning architecture for disease classification.<n>In the second stage, we present a modular report generation pipeline that extracts confidence-weighted diagnostic keywords.
  arXiv  Detail & Related papers  (2025-08-18T16:42:29Z)
• Enhanced Electronic Health Records Text Summarization Using Large Language Models [0.0]
  This project builds on prior work by creating a system that generates clinician-preferred, focused summaries. The proposed system leverages the Flan-T5 model to generate tailored EHR summaries based on clinician-specified topics.
  arXiv  Detail & Related papers  (2024-10-12T19:36:41Z)
• Synchronous Faithfulness Monitoring for Trustworthy Retrieval-Augmented Generation [96.78845113346809]
  Retrieval-augmented language models (RALMs) have shown strong performance and wide applicability in knowledge-intensive tasks. This paper proposes SynCheck, a lightweight monitor that leverages fine-grained decoding dynamics to detect unfaithful sentences. We also introduce FOD, a faithfulness-oriented decoding algorithm guided by beam search for long-form retrieval-augmented generation.
  arXiv  Detail & Related papers  (2024-06-19T16:42:57Z)
• A Data-Centric Approach To Generate Faithful and High Quality Patient Summaries with Large Language Models [11.218649399559691]
  Fine-tuning on hallucination-free data effectively reduces hallucinations from 2.60 to 1.55 per summary for Llama 2. We find that common quantitative metrics do not correlate well with faithfulness and quality.
  arXiv  Detail & Related papers  (2024-02-23T16:32:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.