DEXTER: Diffusion-Guided EXplanations with TExtual Reasoning for Vision Models
This addresses the need for transparent and trustworthy AI systems by providing interpretable explanations without requiring training data, though it is incremental in combining existing techniques for a specific domain.
The paper tackles the problem of explaining visual classifiers by introducing DEXTER, a data-free framework that uses diffusion and language models to generate textual explanations, outperforming existing methods in global model explanation and bias reporting on datasets like ImageNet and CelebA.
Understanding and explaining the behavior of machine learning models is essential for building transparent and trustworthy AI systems. We introduce DEXTER, a data-free framework that employs diffusion models and large language models to generate global, textual explanations of visual classifiers. DEXTER operates by optimizing text prompts to synthesize class-conditional images that strongly activate a target classifier. These synthetic samples are then used to elicit detailed natural language reports that describe class-specific decision patterns and biases. Unlike prior work, DEXTER enables natural language explanation about a classifier's decision process without access to training data or ground-truth labels. We demonstrate DEXTER's flexibility across three tasks-activation maximization, slice discovery and debiasing, and bias explanation-each illustrating its ability to uncover the internal mechanisms of visual classifiers. Quantitative and qualitative evaluations, including a user study, show that DEXTER produces accurate, interpretable outputs. Experiments on ImageNet, Waterbirds, CelebA, and FairFaces confirm that DEXTER outperforms existing approaches in global model explanation and class-level bias reporting. Code is available at https://github.com/perceivelab/dexter.