LANCE: Stress-Testing Visual Models by Generating Language-Guided Counterfactual Images
Abstract
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.
Cite
Text
Prabhu et al. "LANCE: Stress-Testing Visual Models by Generating Language-Guided Counterfactual Images." Neural Information Processing Systems, 2023.Markdown
[Prabhu et al. "LANCE: Stress-Testing Visual Models by Generating Language-Guided Counterfactual Images." Neural Information Processing Systems, 2023.](https://mlanthology.org/neurips/2023/prabhu2023neurips-lance/)BibTeX
@inproceedings{prabhu2023neurips-lance,
title = {{LANCE: Stress-Testing Visual Models by Generating Language-Guided Counterfactual Images}},
author = {Prabhu, Viraj and Yenamandra, Sriram and Chattopadhyay, Prithvijit and Hoffman, Judy},
booktitle = {Neural Information Processing Systems},
year = {2023},
url = {https://mlanthology.org/neurips/2023/prabhu2023neurips-lance/}
}