Search papers, labs, and topics across Lattice.
5
0
7
28
Achieving 1,500 tokens per second, DiffusionGemma redefines the speed-capability trade-off in language models, outpacing conventional autoregressive approaches.
RMMD not only accelerates model inference by 7.5x but also outperforms its teacher model on nearly all target weather variables, showcasing a breakthrough in distillation techniques.
Block verification boosts the efficiency of diffusion models, yielding a surprising 6.3% speedup in inference without additional training.
GMD algorithms, previously seen as a novel generative framework, can be understood as directly targeting fixed points of Wasserstein Gradient Flows, offering a new perspective on their optimization process.
Refining generative models with discriminator guidance provably improves generalization, offering a theoretical justification for techniques like score-based diffusion.
