Search papers, labs, and topics across Lattice.
3
0
2
4
A novel $\mathrm{SemiCat}$ architecture achieves discrete diffusion-level performance on TinyGSM and outperforms one-hot flows at scale by addressing self-conditioning and embedding geometry in continuous diffusion language models.
Masked diffusion language models can match autoregressive baselines while cutting sampling steps by over 20%鈥攑rovided you train across multi-step continuous trajectories rather than isolated random masks.
It is shown that a step matches the training distribution only when the positions it writes are conditionally independent given the tokens already fixed, that no product of per-position distributions can match a dependent group, and that per-position distributions do not determine whether a group is dependent.
