Search papers, labs, and topics across Lattice.
2
3
5
13
Switching between autoregressive and diffusion modes allows Nemotron-Labs-Diffusion to achieve unprecedented throughput and efficiency in language modeling.
Swap out slow, one-token-at-a-time generation in VLMs for a 6x speed boost, without sacrificing quality, using a surprisingly simple direct conversion to block-diffusion decoding.
