2 papers
cs.CV2026
ViTok-v2: Scaling Native Resolution Auto-Encoders to 5 Billion Parameters
Philippe Hansen-Estruch, Jiahui Chen, Vivek Ramanujan +9
Vision Transformer (ViT) autoencoders have emerged as compelling tokenizers for images, offering improved reconstruction over convolutional tokenizers. However, existing ViT tokeni…
cs.CV2026
Interpreting Video Representations with Spatio-Temporal Sparse Autoencoders
Atahan Dokme, Sriram Vishwanath
We present the first systematic study of Sparse Autoencoders (SAEs) on video representations. Standard SAEs decompose video into interpretable, monosemantic features but destroy te…