偏差是子空间，而非坐标：视觉-语言模型中事后去偏的几何重思
Bias Is a Subspace, Not a Coordinate: A Geometric Rethinking of Post-hoc Debiasing in Vision-Language Models
AI总结：
本文提出SPD框架，通过几何方法识别并去除线性可解码的偏子空间，提升视觉-语言模型的公平性与任务性能。
摘要：
视觉-语言模型（VLMs）在多模态推理中不可或缺，但其表示常编码并放大人口偏见，导致下游任务中关联偏差和预测不一致。近期事后方法通过替换最相关的嵌入坐标来减轻偏见，但本文系统分析揭示了该坐标方法的三个关键局限：特征纠缠、跨数据集泛化差和偏见去除不彻底。研究发现偏见并非局限于少数坐标，而是分布在少数线性子空间中。为此，本文提出SPD（Subspace Projection Debiasing）框架，通过几何方法识别并去除整个线性可解码的偏子空间，同时重新插入中性均值成分以保持语义一致性。在零样本分类、文本到图像检索和图像生成等任务中，SPD经过广泛实验验证：在四个公平性指标上平均提升18.5%，同时相比最佳去偏基线保持任务性能损失最小。
英文摘要：
Vision-Language Models (VLMs) have become indispensable for multimodal reasoning, yet their representations often encode and amplify demographic biases, resulting in biased associations and misaligned predictions in downstream tasks. Such behavior undermines fairness and distorts the intended alignment between vision and language. Recent post-hoc approaches attempt to mitigate bias by replacing the most attribute-correlated embedding coordinates with neutral values. However, our systematic analysis reveals three critical limitations of this coordinate-wise approach: feature entanglement, poor cross-dataset generalization, and incomplete bias removal. We find that bias is not localized to a few coordinates but is instead distributed across a few linear subspaces. To address these limitations, we propose $\textbf{S}$ubspace $\textbf{P}$rojection $\textbf{D}$ebiasing ($\textbf{SPD}$), a geometrically principled framework that identifies and removes the entire subspace of linearly decodable bias while reinserting a neutral mean component to preserve semantic fidelity. Extensive experiments across zero-shot classification, text-to-image retrieval, and image generation validate the effectiveness of SPD: our method achieves more robust debiasing with an average improvement of $18.5\%$ across four fairness metrics, while maintaining minimal loss in task performance compared to the best debiasing baseline.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
