用于医学视觉语言模型交叉公平性的双曲关系提示词
Hyperbolic Relational Prompts for Intersectional Fairness in Medical VLMs
把提示词生成建模为在双曲空间中基于样本级关系图的推理过程，为视觉语言模型提供了一种可复用的条件化手段，可用于处理交叉属性组合（如种族×性别）下的细粒度偏差；做提示词微调或多属性可控生成时，可借鉴其双曲图层的层级建模思路。
原文依据 FRP constructs a relational graph to capture fine-grained, sample-level similarities and employs a hyperbolic graph layer to explicitly model the hierarchical structure of intersectional identities.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论确保医学视觉语言模型（VLM）的公平性对于实现公平医疗至关重要，然而现有模型会放大种族、性别等人口学子群之间的偏差。依赖宽泛分布对齐的传统公平性缓解方法，难以应对这些细微的交叉性差异。我们提出公平感知关系提示（FRP），这是一个将提示词生成重新表述为动态、公平感知推理过程的新框架。FRP 构建关系图以捕捉细粒度的样本级相似性，并采用双曲图图层显式建模交叉身份的层级结构。利用双曲几何可对复杂属性组合进行推理，从而有效减少根深蒂固的偏差。在 FairVLMed 和 Harvard-GF 数据集上的评估表明，FRP 取得了最优的诊断性能，曲线下面积分别为 77.50% 和 85.94%，同时显著改善了人口统计学均等差异和均衡几率差异。
英文原文摘要 1101 字符
Ensuring fairness in medical vision-language models (VLMs) is essential for equitable healthcare, yet existing models amplify biases across demographic subgroups such as race and gender. Traditional fairness mitigation approaches relying on broad distribution alignment, fall short in addressing these nuanced intersectional disparities. We propose fairness-aware relational prompting (FRP), a novel framework that reformulates prompt generation as a dynamic, fairness-aware reasoning process. FRP constructs a relational graph to capture fine-grained, sample-level similarities and employs a hyperbolic graph layer to explicitly model the hierarchical structure of intersectional identities. Leveraging hyperbolic geometry enables reasoning over complex attribute combinations, effectively reducing entrenched biases. Evaluations on the FairVLMed and Harvard-GF datasets demonstrate that FRP achieves state-of-the-art diagnostic performance, with an area under the curve of 77.50% and 85.94% respectively, while substantially improving the demographic parity difference and equalized odds difference.