SafeFix: 通过受控图像生成实现目标模型修复
SafeFix: Targeted Model Repair via Controlled Image Generation
AI总结：
SafeFix通过生成语义忠实的图像来修复模型，提升模型对罕见案例的鲁棒性，减少系统性错误。
摘要：
深度学习模型在视觉识别中常因语义子群体不足而出现系统性错误。尽管现有的调试框架可以通过识别关键失败属性来定位这些失败，但有效修复模型仍然具有挑战性。当前的解决方案通常依赖于人工设计的提示来生成合成训练图像——这种方法容易导致分布偏移和语义错误。为克服这些挑战，我们引入了一个模型修复模块，该模块基于可解释的失败归因流程。我们的方法使用条件文本到图像模型生成语义忠实且针对性强的图像用于失败案例。为了保持生成样本的质量和相关性，我们进一步采用大型视觉-语言模型（LVLM）来过滤输出，强制对齐原始数据分布并保持语义一致性。通过使用此罕见案例增强的合成数据集重新训练视觉模型，我们显著减少了与罕见案例相关的错误。我们的实验表明，这种针对性修复策略在不引入新错误的情况下提高了模型的鲁棒性。
英文摘要：
Deep learning models for visual recognition often exhibit systematic errors due to underrepresented semantic subpopulations. Although existing debugging frameworks can pinpoint these failures by identifying key failure attributes, repairing the model effectively remains difficult. Current solutions often rely on manually designed prompts to generate synthetic training images -- an approach prone to distribution shift and semantic errors. To overcome these challenges, we introduce a model repair module that builds on an interpretable failure attribution pipeline. Our approach uses a conditional text-to-image model to generate semantically faithful and targeted images for failure cases. To preserve the quality and relevance of the generated samples, we further employ a large vision-language model (LVLM) to filter the outputs, enforcing alignment with the original data distribution and maintaining semantic consistency. By retraining vision models with this rare-case-augmented synthetic dataset, we significantly reduce errors associated with rare cases. Our experiments demonstrate that this targeted repair strategy improves model robustness without introducing new bugs. Code is available at https://github.com/oxu2/SafeFix
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。
