A Benchmark for Hallucination Detection in VLMs for Gastrointestinal Endoscopy

This study introduces the Gut-VLM dataset, a benchmark for hallucination detection in vision-language models (VLMs) specifically for gastrointestinal endoscopy, comprising 4,392 test VQA pairs evaluated across five models: MedGemma-4B, MedGemma-27B, LLaVA-Med-7B, LLaVA-v1.6-7B, and Lingshu-32B. The evaluation of nine detection methods reveals that ReXTrust, a white-box method, achieves the highest average AUC of 93.0 on MedGemma-4B, significantly outperforming alternatives, while highlighting the challenge of "confident confabulation" as a common failure mode. This benchmark is crucial for practitioners as it addresses the safety concerns of deploying VLMs in clinical settings, providing insights into effective detection strategies.