arXiv cs.AI· Sachin Acharya, Niyoj Oli, Shishir Adhikari, Aminu Lawal, Yash Raj Shrestha, Prashnna K Gyawali, Binod Bhattarai·· 3 小时前AI 评分33
MAGIC：面向 LMM 临床幻觉检测的模态感知基尼系数检查方法
MAGIC: Modality-Aware Gini Inspection for Clinical Hallucination Detection in LMMs
AI 导读
论文提出 MAGIC 方法，把医疗多模态模型（LMM）的幻觉检测建模为生成答案 token 时模态间注意力权重分布失衡问题。该方法用基尼系数量化各层、各注意力头中视觉与文本 token 的注意力集中度，发现幻觉回答中注意力质量存在统计显著的重分布。基于这些特征训练的逻辑回归分类器在两个医疗 LMM、三种医学影像模态上评估，表现达到或超过现有幻觉检测基线。
正文 · 原文
Abstract:Recent advances in medical large multimodal models (LMMs) have brought promising progress to clinical decision support, yet hallucination, the generation of confident but clinically unverifiable outputs, poses a critical threat to the reliability demanded in high-stakes medical settings. Existing hallucination detection methods, predominantly designed for unimodal settings, fail to capture the visual-linguistic dynamics inherent to medical imaging. We formulate the problem of hallucination detection in LMMs as disproportionate distribution of attention weights across the modalities during answer token generation. We quantify modality-specific attention concentration across layers and heads using the Gini coefficient and demonstrate statistically significant redistribution of attention mass between visual and textual tokens in hallucinated responses. We leverage these features to train a logistic regression-based classifier and evaluate it on two medical LMMs across three medical imaging modalities, demonstrating that our method achieves competitive or superior performance against contemporary hallucination detection baselines.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.15530 [cs.CV] |
| (or arXiv:2611.15530v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2611.15530 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Sachin Acharya [view email]
[v1]
Wed, 11 Nov 2026 14:16:43 UTC (813 KB)
来源：arXiv cs.AI · arxiv.org