Multi-Block Diffusion Language Models

Multi-Block Diffusion Language Models

Multi-Block Diffusion Language Models extend single-block diffusion to concurrent block decoding with improved training strategies and optimized decoding algorithms.

Hugging Face · Daily Papers ·Yijie Jin, Jiajun Xu · ·▲ 22 upvotes

Este artigo está em destaque na seleção diária de papers do Hugging Face, curada pela comunidade de pesquisa em IA.

Autores: Yijie Jin, Jiajun Xu, Yuxuan Liu, Chenkai Xu, Yi Tu, Jiajun Li

  • 22 upvotes da comunidade
  • Temas: Block Diffusion Language Models, Multi-Block Diffusion, teacher forcing, diffusion forcing, Multi-block Teacher Forcing, noise-schedulers

Resumo

Resumo original (em inglês), extraído do paper:

Multi-Block Diffusion Language Models extend single-block diffusion to concurrent block decoding with improved training strategies and optimized decoding algorithms.

Onde ler

compartilhar: