Search papers, labs, and topics across Lattice.
4
0
8
3
SCOPE achieves a remarkable 1.99脳 speedup in video attention while enhancing fidelity, challenging the effectiveness of traditional sparse attention methods.
Structural template tokens in diffusion transformers act as implicit semantic registers, crucial for maintaining object identity despite carrying minimal prompt-specific information.
Breaking the tight coupling between feature engineering and model architecture, PG enables rapid iterations and deployment in generative retrieval systems, yielding measurable performance gains.
Stop treating diffusion workflows as monolithic black boxes: LegoDiffusion unlocks 3x higher throughput by decomposing them into independently scalable microservices.
