arXiv cs.AI· Jian Yang, Peilin Liu, Xiaohan Li, Songda Li, Yu Cao, Yuqing Feng, Haomin Li·· 2 小时前AI 评分50
审计大语言模型罕见病诊断评测中的数据泄漏与评测缺陷
Auditing evaluation artifacts and data leakage in large language model systems for rare disease diagnosis
AI 导读
一项实证审计研究基于超过 60,000 例临床病例，量化了增强型大语言模型（LLM）罕见病诊断评测中的数据泄漏问题。重建先进 agentic 框架后发现，严格跨数据集隔离使平均 Recall@1 从 54.3% 降至 40.3%（Hedges' g = 0.94），监督分类器外部准确率从 42.9% 跌至 16.7%。
正文 · 原文
Abstract:Augmented large language model (LLM) systems have reported high benchmark accuracies in rare disease diagnosis, yet their real-world clinical utility remains unproven. A recent meta-analysis flagged augmented LLM diagnostic evaluations as having a high risk of bias due to suspected data leakage; however, the precise impact of these vulnerabilities on model inference has yet to be fully quantified. In this study, we present a comprehensive empirical audit to quantify evaluation artifacts across prevailing AI paradigms, using a repository of over 60,000 cases from diverse public and independent clinical cohorts. By reconstructing a state-of-the-art agentic framework, we demonstrate that candidate-space contamination artificially inflates performance: enforcing strict cross-dataset isolation resulted in a marked performance decline, with the cross-dataset mean Recall@1 decreasing from 54.3% (under conventional one-vs-rest evaluation) to 40.3% (Hedges' g = 0.94). Similar spatial overfitting in supervised classifiers caused diagnostic accuracy to drop from an internal mean of 42.9% to 16.7% across external cohorts (Cohen's d = 2.01). Furthermore, we quantify how LLMs infer diagnoses by exploiting retrospective therapeutic cues in clinical narratives, and how temporal misalignment in foundational knowledge bases leads to performance fluctuations. We conclude that prevailing evaluation protocols for augmented LLM systems inadvertently reflect data leakage rather than genuine medical reasoning. To address these systematic artifacts, we propose a rigorous evaluation framework - coupling strict data-isolation protocols with standardized benchmarking safeguards - to ensure future diagnostic AI for rare diseases is evaluated on actual clinical robustness.
| Comments: | 18 page, 3 figures |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.33592 [cs.AI] |
| (or arXiv:2610.33592v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33592 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Haomin Li [view email]
[v1]
Sat, 24 Oct 2026 10:35:48 UTC (555 KB)
来源：arXiv cs.AI · arxiv.org