From Next-Token to Next-Block: A Principled Adaptation Path for Diffusion LLMs
Paper
•
2512.06776
•
Published
•
24
YuchuanTian
yuchuantian
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
18 days ago
VersatileFFN: Achieving Parameter Efficiency in LLMs via Adaptive Wide-and-Deep Reuse
published
a Space
25 days ago
yuchuantian/NBDiff
upvoted
a
paper
25 days ago
From Next-Token to Next-Block: A Principled Adaptation Path for Diffusion LLMs
Organizations
None yet
submitted
a
paper
to Daily Papers
25 days ago
authored
a
paper
over 1 year ago