From Next-Token to Next-Block: A Principled Adaptation Path for Diffusion LLMs
Paper
•
2512.06776
•
Published
•
24
Chengxuan Zhu
zcx65535
AI & ML interests
Computer Vision, Computational Photography
Recent Activity
upvoted
a
paper
21 days ago
From Next-Token to Next-Block: A Principled Adaptation Path for Diffusion LLMs
updated
a model
6 months ago
zcx65535/bokehdiff
published
a model
6 months ago
zcx65535/bokehdiff
Organizations
None yet