Author

Mikhail Dereviannykh

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#machine learning Preprint Sep 2026

SemanTok: Predictable Semantic Tokens for Efficient Autoregressive Video Generation

SemanTok is introduced, a flexible video tokenizer that feeds frozen DINO features into its encoder and adds lightweight heads that reconstruct them from each retained token prefix alone, and performs well in both reconstruction and generation.

Mikhail Dereviannykh, Vikram Voleti, Simon Donné et al. · 0 citations