Search papers, labs, and topics across Lattice.
3
0
5
2
Forget billion-scale datasets: EvoTok achieves state-of-the-art image tokenization for both understanding and generation using a residual evolution process trained on just 13M images.
Current image editing models struggle with discipline-specific knowledge, failing to consistently produce edits that are both visually consistent and logically sound within academic domains.
DreamWorld achieves more world-consistent video generation by jointly modeling multiple heterogeneous dimensions of world knowledge, moving beyond surface-level plausibility.
