HalluCXR: Benchmarking and Mitigating Hallucinations in Medical Vision-Language Models for Chest Radiograph Interpretation

• URL: http://arxiv.org/abs/2605.20469v1
• Date: Tue, 19 May 2026 20:30:32 GMT
• Title: HalluCXR: Benchmarking and Mitigating Hallucinations in Medical Vision-Language Models for Chest Radiograph Interpretation
• Abstract summary: HalluCXR is a benchmark evaluating six architecturally diverse vision-language models across 856 chest radiographs.<n>We find that 61.9--82.3% of outputs contain hallucinations, with clinically dangerous errors in up to 80.2%.<n>These results establish that hallucination auditing, verbosity-based risk monitoring, and ensemble-based safety layers are prerequisites for clinical deployment.
• Score: 7.512887610816016
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vision-language models (VLMs) are increasingly used for medical image interpretation, yet they frequently hallucinate, generating clinically plausible but factually incorrect findings that pose direct patient safety risks. We introduce HalluCXR, a benchmark evaluating six architecturally diverse VLMs across 856 stratified MIMIC-CXR chest radiographs and three query types, yielding 15,408 model evaluations. An eight-category hallucination taxonomy with clinical severity ratings and a two-layer detection pipeline are validated against 250 human annotations (auto-detection F1=0.959; LLM judge F1=0.907). We find that 61.9--82.3% of outputs contain hallucinations, with clinically dangerous errors in up to 80.2%. Three key patterns emerge: normal radiographs paradoxically attract the most severe hallucinations, common findings are systematically over-fabricated while rare findings go under-detected, and response length alone predicts hallucination risk (AUC up to 0.908). A six-model ensemble reduces fabrication by up to 84.8% at the cost of increased omission; a three-model subset retains comparable performance at half the cost. These results establish that hallucination auditing, verbosity-based risk monitoring, and ensemble-based safety layers are prerequisites for clinical deployment.

Related papers

• RareLens: Towards End-to-End Rare Disease Care via Aligning Divergent Large Language Model Reasoning [42.80742579670377]
  RareLens is a system that supports clinical decision-making across the entire rare disease trajectory.<n>Developed and evaluated on RareBench, a real-world dataset of 157,525 cases spanning all 33 Orphanet categories and more than 7,000 conditions.<n>It achieved an area under the curve of 0.917 for screening and top-1 accuracies of 65.5% and 89.8% for diagnosis and treatment.
  arXiv  Detail & Related papers  (2026-07-25T16:58:10Z)
• HalluScan: A Systematic Benchmark for Detecting and Mitigating Hallucinations in Instruction-Following LLMs [0.0]
  Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse natural language processing tasks.<n>Yet they remain susceptible to hallucinations -- generating content that is factually incorrect, unfaithful to provided context, or misaligned with user instructions.<n>We present HalluScan, a comprehensive benchmark framework that systematically evaluates hallucination detection and mitigation across 72 configurations.
  arXiv  Detail & Related papers  (2026-05-04T10:43:27Z)
• GAZE: Grounded Agentic Zero-shot Evaluation with Viewer-Level Tools and Literature Retrieval on Rare Brain MRI [0.0]
  Vision-language models (VLMs) read an image and produce text in a single forward pass, whereas radiologists typically inspect an image several times and consult the literature before writing a report.<n>We introduce GAZE, a framework that lets a medical VLM work in this iterative way by calling viewer-level tools and two retrieval tools backed by the U.S. National Library of Medicine.<n>On NOVA, a benchmark of 906 brain MRI cases covering 281 rare neurological conditions, GAZE reaches 58.2 mean average precision (mAP) at intersection-over-union (IoU) 0.3 for lesion localisation and
  arXiv  Detail & Related papers  (2026-04-25T13:38:49Z)
• Medical thinking with multiple images [44.04557445622649]
  We introduce MedThinkVQA, an expert-annotated benchmark for thinking with multiple images.<n>The dataset contains 8,067 cases, including 720 test cases, with an average of 6.62 images per case.<n>On the test set, the best closed-source models, Claude-4.6-Opus, Gemini-3-Pro, and GPT-5.2-xhigh, reach only 57.2%, 55.3%, and 54.9% accuracy.
  arXiv  Detail & Related papers  (2026-04-14T18:51:07Z)
• PanCanBench: A Comprehensive Benchmark for Evaluating Large Language Models in Pancreatic Oncology [48.732366302949515]
  Large language models (LLMs) have achieved expert-level performance on standardized examinations, yet multiple-choice accuracy poorly reflects real-world clinical utility and safety.<n>We developed a human-in-the-loop pipeline to create expert rubrics for de-identified patient questions.<n>We evaluated 22 proprietary and open-source LLMs using an LLM-as-a-judge framework, measuring clinical completeness, factual accuracy, and web-search integration.
  arXiv  Detail & Related papers  (2026-03-02T00:50:39Z)
• Suppressing Prior-Comparison Hallucinations in Radiology Report Generation via Semantically Decoupled Latent Steering [94.37535002230504]
  We develop a training-free, inference-time control framework termed Semantically Decoupled Latent Steering.<n>Our approach constructs a semantic-free intervention vector via large language model (LLM)-driven semantic decomposition.<n>We show that our approach significantly reduces the probability of historical hallucinations.
  arXiv  Detail & Related papers  (2026-02-27T04:49:01Z)
• Diagnosing Hallucination Risk in AI Surgical Decision-Support: A Sequential Framework for Sequential Validation [5.469454486414467]
  Large language models (LLMs) offer transformative potential for clinical decision support in spine surgery.<n>LLMs pose significant risks through hallucinations, which are factually inconsistent or contextually misaligned outputs.<n>This study introduces a clinician-centered framework to quantify hallucination risks by evaluating diagnostic precision, recommendation quality, reasoning robustness, output coherence, and knowledge alignment.
  arXiv  Detail & Related papers  (2025-11-01T15:25:55Z)
• 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)
• Trustworthy AI for Medicine: Continuous Hallucination Detection and Elimination with CHECK [1.3638020767676653]
  Large language models (LLMs) show promise in healthcare, but hallucinations remain a major barrier to clinical use.<n>We present CHECK, a continuous-learning framework that integrates structured clinical databases to detect hallucinations.
  arXiv  Detail & Related papers  (2025-06-10T17:12:28Z)
• Medical Hallucinations in Foundation Models and Their Impact on Healthcare [71.15392179084428]
  Hallucinations in foundation models arise from autoregressive training objectives.<n>Top-performing models exceeded 97% accuracy when augmented with chain-of-thought prompting.
  arXiv  Detail & Related papers  (2025-02-26T02:30:44Z)
• MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models [82.30696225661615]
  We introduce MedHallu, the first benchmark specifically designed for medical hallucination detection.<n>We show that state-of-the-art LLMs, including GPT-4o, Llama-3.1, and medically fine-tuned UltraMedical, struggle with this binary hallucination detection task.<n>Using bidirectional entailment clustering, we show that harder-to-detect hallucinations are semantically closer to ground truth.
  arXiv  Detail & Related papers  (2025-02-20T06:33:23Z)
• CXR-LLAVA: a multimodal large language model for interpreting chest X-ray images [3.0757789554622597]
  This study aimed to develop an open-source multimodal large language model (CXR-LLAVA) for interpreting chest X-ray images (CXRs) For training, we collected 592,580 publicly available CXRs, of which 374,881 had labels for certain radiographic abnormalities. The model's diagnostic performance for major pathological findings was evaluated, along with the acceptability of radiologic reports by human radiologists.
  arXiv  Detail & Related papers  (2023-10-22T06:22:37Z)

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.