Deep learning models are notoriously opaque. Existing explanation methods often focus on localized visual explanations for individual images. Concept-based explanations, while offering global insights, require extensive annotations, incurring significant labeling cost. We propose an approach that leverages user-defined part labels from a limited set of images and efficiently transfers them to a larger dataset. This enables the generation of global symbolic explanations by aggregating part-based local explanations, ultimately providing human-understandable explanations for model decisions on a large scale.
@article{arxiv.2509.15393,
title = {Generating Part-Based Global Explanations Via Correspondence},
author = {Kunal Rathore and Prasad Tadepalli},
journal= {arXiv preprint arXiv:2509.15393},
year = {2025}
}