发现临床语言模型会放大种族性别偏见，而非简单复制训练数据。
A Computational Audit of Demographic Association Encoding in ClinicalBERT Language Predictions
- 用概率偏差分析和掩码语言模型探针检测模型内部偏见
- 65.6%的显著结果与真实病历数据相反，黑人患者高达80%
- 偏见主要来自模型内部放大，非训练数据直接继承
基于Transformer的临床语言模型正被广泛应用于高风险医疗决策支持系统，但其在医学记录中编码的人口统计学关联如何影响模型输出概率分布，仍缺乏实证研究。本文对ClinicalBERT（Alsentzer等，2019）进行系统性计算审计，该模型基于MIMIC-III出院记录预训练。采用两种互补探针方法：对数概率偏差分析（LPBA），量化不同人口特征导致的掩码词概率分布变化；掩码语言模型分析（MLM），探测98个真实临床句式模板中8种交叉种族-性别组合下代理权归属的内部表征。通过语料库频率分析，区分统计差异与偏见放大。32个显著发现中，65.6%与原始语料分布矛盾，其中黑人患者相关结果达80%，代理权归属分析下更达87.5%，表明ClinicalBERT中的表征偏见主要源于模型内部放大机制，而非训练数据直接继承。
原文摘要 · Abstract (English)
Transformer-based clinical language models are increasingly integrated into high-stakes clinical decision support pipelines, yet the computational mechanisms through which demographic associations encoded in medical documentation propagate into model probability distributions remain empirically underspecified. We present a systematic computational audit of representational bias in ClinicalBERT (Alsentzer et al., 2019), a BERT-based model pretrained on MIMIC-III discharge summaries, employing two complementary probing methodologies: Log Probability Bias Analysis (LPBA), which quantifies demographic descriptor-induced shifts in masked token probability distributions across behavioral and evaluative semantic categories, and Masked Language Model-based analysis (MLM), which probes internal representational structure for demographic agency attribution encoding across 98 real clinical sentence templates and eight intersectional race-gender combinations. Corpus frequency analysis operationalizes the distinction between statistical disparity and bias amplification by benchmarking model outputs against empirical term frequencies in the MIMIC-III training corpus. Of 32 statistically significant findings, 65.6% contradict observed corpus distributions, rising to 80% for Black patients and 87.5% for agency attribution under MLM probing, providing direct empirical evidence that representational bias in ClinicalBERT operates predominantly through model-internal amplification rather than training data inheritance. Keywords: natural language processing, clinical documentation, algorithmic auditing, representational bias, health equity 1
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。