Search papers, labs, and topics across Lattice.
4
0
8
6
MLLMs possess an intrinsic "S-Space" that acts as a spatial workspace, encoding object coordinates and enabling viewpoint transformations, even under non-spatial queries.
HarnessEval-W transforms world model evaluation from mere scoring to a transparent reasoning process that mirrors human judgment.
TivTok redefines video tokenization by enabling the reuse of persistent information across frames, achieving unprecedented compression efficiency with minimal token usage.
Generative videos might look great, but a new metric reveals they often suffer from jarring 3D spatial inconsistencies that existing metrics miss.
