EH-Benchmark Ophthalmic Hallucination Benchmark and Agent-Driven Top-Down Traceable Reasoning Workflow
On the map: LLM reasoning → Hallucination detection and mitigation
also in 1 more
Abstract (arXiv)
Medical Large Language Models (MLLMs) play a crucial role in ophthalmic diagnosis, holding significant potential to address vision-threatening diseases. However, their accuracy is constrained by hallucinations stemming from limited ophthalmic knowledge, insufficient visual localization and reasoning capabilities, and a scarcity of multimodal ophthalmic data, which collectively impede precise lesion detection and disease diagnosis. Furthermore, existing medical benchmarks fail to effectively evaluate various types of hallucinations or provide actionable solutions to mitigate them. To address the above challenges, we introduce EH-Benchmark, a novel ophthalmology benchmark designed to evaluate hallucinations in MLLMs. We categorize MLLMs' hallucinations based on specific tasks and error types into two primary classes: Visual Understanding and Logical Composition, each comprising multiple subclasses. Given that MLLMs predominantly rely on language-based reasoning rather than visual processing, we propose an agent-centric, three-phase framework, including the Knowledge-Level Retrieval stage, the Task-Level Case Studies stage, and the Result-Level Validation stage. Experimental results show that our multi-agent framework significantly mitigates both types of hallucinations, enhancing accuracy, interpretability, and reliability. Our project is available at https://github.com/ppxy1/EH-Benchmark.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Hallucination detection and mitigation, one of 17 directions in LLM reasoning. Builds on 29 of its 69 references that are in the map. First posted on arXiv 24 Jul 2025.
Builds on
29 of 69 references.
References · 29
- 2023GPT-4 Technical Reportcited by 18,132
- 2025Qwen2.5-VL Technical Reportcited by 5,418
- 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,151
- 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,179
- 2024Qwen2 Technical Reportcited by 2,076
- 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contextscited by 1,582
- 2024Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scalingcited by 1,573
- 2024How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suitescited by 1,033
- 2024DeepSeek-V3 Technical Reportcited by 724
- 2023Multi-Agent Collaboration: Harnessing the Power of Intelligent LLM Agentscited by 329
- 2024Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimizationcited by 238
- 2024HuatuoGPT-Vision, Towards Injecting Medical Visual Knowledge into Multimodal LLMs at Scalecited by 194
17 more
- 2025HealthGPT: A Medical Large Vision-Language Model for Unifying Comprehension and Generation via Heterogeneous Knowledge Adaptationcited by 112
- 2024Mini-InternVL: A Flexible-Transfer Pocket Multimodal Model with 5% Parameters and 90% Performancecited by 94
- 2024Embodied LLM Agents Learn to Cooperate in Organized Teamscited by 81
- 2024Detecting and Evaluating Medical Hallucinations in Large Vision Language Modelscited by 57
- 2025LLM Agents Making Agent Toolscited by 45
- 2023ChatKBQA: A Generate-then-Retrieve Framework for Knowledge Base Question Answering with Fine-tuned Large Language Modelscited by 44
- 2025LLMEval-Med: A Real-world Clinical Benchmark for Medical LLMs with Physician Validationcited by 29
- 2025EyecareGPT: Boosting Comprehensive Ophthalmology Understanding with Tailored Dataset, Benchmark and Modelcited by 22
- 2024MedVH: Towards Systematic Evaluation of Hallucination for Large Vision Language Models in the Medical Contextcited by 19
- 2024MedSAM-U: Uncertainty-Guided Auto Multi-Prompt Adaptation for Reliable MedSAMcited by 17
- 2024MedHallBench: A New Benchmark for Assessing Hallucination in Medical Large Language Modelscited by 15
- 2025Self-supervised Quantized Representation for Seamlessly Integrating Knowledge Graphs with Large Language Modelscited by 11
- 2025MedHEval: Benchmarking Hallucinations and Mitigation Strategies in Medical Large Vision-Language Modelscited by 10
- 2024Measuring the Measurers: Quality Evaluation of Hallucination Benchmarks for Large Vision-Language Modelscited by 8
- 2024LMOD: A Large Multimodal Ophthalmology Dataset and Benchmark for Large Vision-Language Modelscited by 7
- 2025FunBench: Benchmarking Fundus Reading Skills of MLLMscited by 5
- 2025An integrated language-vision foundation model for conversational diagnostics and triaging in primary eye carecited by 3
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2026Med-StepBench: A Hierarchical Reasoning Framework for Evaluating Hallucinations in Medical Vision-Language Models
- 2025MedHallTune: An Instruction-Tuning Benchmark for Mitigating Medical Hallucination in Vision-Language Models
- 2026ClinHallu: A Benchmark for Diagnosing Stage-Wise Hallucinations in Medical MLLM Reasoning
- 2026Synergistic Perception-Reasoning Governance: Grounding Medical MLLMs with Verifiable Anatomical Evidence
- 2026HalluScope: Fine-grained Hallucination Diagnosis for Multimodal Large Language Models
- 2026EyeVQA: Benchmarking Ophthalmic Vision-Language Models from Recognition to Spatial Grounding
- 2026HalluGuard: Demystifying Data-Driven and Reasoning-Driven Hallucinations in LLMs
- 2026Deterministic Hallucination Detection in Medical VQA via Confidence-Evidence Bayesian Gain