3 papers
cs.CV2026
Beyond Interpretability: When, Why, and How Sparse Autoencoders Enable Label-Free Visual Steering
Gerasimos Chatzoudis, Zhuowei Li, Gemma E. Moran +2
Sparse Autoencoders (SAEs) are increasingly used to interpret foundation models, but their role as an actionable intervention space remains less understood, especially in vision. W…
cs.CV2026
Can Cross-Layer Transcoders Replace Vision Transformer Activations? An Interpretable Perspective on Vision
Gerasimos Chatzoudis, Konstantinos D. Polyzos, Zhuowei Li +4
Understanding the internal activations of Vision Transformers (ViTs) is critical for building interpretable and trustworthy models. While Sparse Autoencoders (SAEs) have been used…
cs.CV2026
LUCID-SAE: Learning Unified Vision-Language Sparse Codes for Interpretable Concept Discovery
Difei Gu, Yunhe Gao, Gerasimos Chatzoudis +6
Sparse autoencoders (SAEs) offer a natural path toward comparable explanations across different representation spaces. However, current SAEs are trained per modality, producing dic…