Breakthroughs, discoveries, and DIY tips sent every weekday. Observing animals, especially other social primates, can be awe-inspiring. Seeing non-human species groom, feed, or socialize with their friends and kin echoes the best of our own impulses. It can feel affirming to know that, in many ways, they're like us. But, like humans, other primates are complicated.

We propose an automated algorithm to stress-test a trained visual model by generating language-guided counterfactual test images (LANCE). Our method leverages recent progress in large language modeling and text-based image editing to augment an IID test set with a suite of diverse, realistic, and challenging test images without altering model weights. We benchmark the performance of a diverse set of pre-trained models on our generated data and observe significant and consistent performance drops. We further analyze model sensitivity across different types of edits, and demonstrate its applicability at surfacing previously unknown class-level model biases in ImageNet. Code is available at https://github.com/virajprabhu/lance.