Not Just Pretty Pictures: Toward Interventional Data Augmentation Using Text-to-Image Generators

Neural image classifiers are known to undergo severe performance degradation when exposed to inputs that exhibit covariate shifts with respect to the training distribution. A general interventional data augmentation (IDA)mechanism that simulates arbitrary interventions over spurious variables has often been conjectured as a theoretical solution to this problem and approximated to varying degrees of success. In this work, we study how well modern Text-to-Image (T2I) generators and associated image editing techniques can solve the problem of IDA. We experiment across a diverse collection of benchmarks in domain generalization, ablating across key dimensions of T2I generation, including interventional prompts, conditioning mechanisms, and post-hoc filtering, showing that it substantially outperforms previously state-of-the-art image augmentation techniques independently of how each dimension is configured. We discuss the comparative advantages of using T2I for image editing versus synthesis, also finding that a simple retrieval baseline presents a surprisingly effective alternative, which raises interesting questions about how generative models should be evaluated in the context of domain generalization.