发表机构
Télécom Paris; Institut Polytechnique de Paris（巴黎电信学院; 巴黎综合理工学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出基于对比分析（CA）的无分类器视觉反事实生成方法，利用StyleGAN2和特征空间F，在三个医学成像数据集上实现了更优的反事实生成质量。
AI 中文摘要
视觉反事实解释（VCEs）旨在通过生成输入图像的最小编辑且真实的版本来改变分类器的预测，从而解释图像分类器。现有VCE方法本质上依赖分类器，因此易受分类器偏差和失效模式影响，例如对捷径特征敏感及校准误差。本文提出一种基于对比分析（CA）的无分类器视觉反事实生成方法。给定对应不同类别的两个数据集（如健康人群与患者），我们将两个数据集共有的生成因子与各数据集特有的显著因子解耦，仅交换显著因子生成反事实图像。由于直接作用于数据分布而非决策边界，该方法提供模型不可知的VCE，对分类器偏差敏感性更低。我们利用StyleGAN2的高质量合成能力与结构良好的隐空间，采用特征空间F而非常规W空间以提升细节保留度。传统CA方法通常假设仅一个数据集存在显著因子，为此我们引入适配框架与损失函数，使各数据集可存在多个显著因子。在三个医学成像数据集上的评估表明，与现有方法相比，本文方法的反事实生成质量更优。
英文摘要
Visual Counterfactual Explanations (VCEs) aim to explain image classifiers by generating minimally edited and realistic versions of an input image that change the classifier's prediction. Existing VCE methods are inherently classifier-dependent and therefore susceptible to classifier biases and failure modes, such as sensitivity to shortcut features and calibration errors. In this paper, we propose a classifier-free approach for visual counterfactual generation based on Contrastive Analysis (CA). Given two datasets corresponding to different classes (e.g., healthy and patients), we disentangle the generative factors that are common across the two datasets from those that are salient to each dataset, and generate counterfactual images by swapping only the salient factors. By operating directly on data distributions rather than decision boundaries, our method provides model-agnostic VCEs that are less sensitive to classifier biases. Our approach leverages the high-quality synthesis and well-structured latent space of StyleGAN2. We use the feature space F, instead than the usual W-space, to improve detail preservation. Unlike conventional CA approaches, which typically assume salient factors in only one dataset, we introduce an adapted framework and loss functions for VCE that allow multiple salient factors in each dataset. We evaluate our method on three medical imaging datasets and demonstrate superior counterfactual generation quality compared to existing approaches.
CommentsMICCAI 2026
