Code Vulnerability Detection via Signal-Aware Learning
Sahil Suneja, Yufan Zhuang, et al.
EuroS&P 2023
Learning visual representations with interpretable features, i.e., disentangled representations, remains a challenging problem. Existing methods demonstrate some success but are hard to apply to large-scale vision datasets like ImageNet. In this work, we propose a simple post-processing framework to disentangle content and style in learned representations from pre-trained vision models. We model the pre-trained features probabilistically as linearly entangled combinations of the latent content and style factors and develop a simple disentanglement algorithm based on the probabilistic model. We show that the method provably disentangles content and style features and verify its efficacy empirically. Our post-processed features yield significant domain generalization performance improvements when the distribution shift occurs due to style changes or style-related spurious correlations.
Sahil Suneja, Yufan Zhuang, et al.
EuroS&P 2023
Kohei Miyaguchi, Masao Joko, et al.
ASMC 2025
Mateo Espinosa Zarlenga, Gabriele Dominici, et al.
ICML 2025
Alec Helbling, Tuna Meral, et al.
ICML 2025