Search papers, labs, and topics across Lattice.
Affiliation:
3
0
5
Uncertainty DMD is proposed, a simple uncertainty-injection framework that restores stochasticity at two key stages of AR generation: a timestep perturbation for the first chunk to increase first-chunk diversity, and a stochastic cache-writing mechanism for later chunks to preserve uncertainty in autoregressive conditioning.
Current image editing models struggle with physics-based reasoning, as revealed by the new PhyEditBench benchmark.
GIM-World achieves superior long-horizon visual consistency by integrating geometry-aware implicit memory, outperforming traditional memory systems.
