Product
Help
Pricing
About
Jobs
Log in
Get started
Author
Add to collection
Haofeng Xu
Observed name:
Xu, Haofeng
Institutions:
Alibaba Group (China),
University of Hong Kong
Observed institutions:
Alibaba Group (China),
University of Hong Kong
OpenAlex ID:
A5128022395
GAC: Stabilizing Asynchronous RL Training for LLMs via Gradient Alignment Control
2026
·
Xu, Haofeng
, Su, Junwei
, et al.
·
arXiv (Cornell University)
·
5
PDF
Scaling Laws for Collapse in Asynchronous GRPO
2026
·
Song, Jingwei
, Xu, Haofeng
, et al.
·
arXiv (Cornell University)
·
4
PDF
ReFlow: Replay-Driven Decoupling for Efficient RL Post-Training
2026
·
Tian, Yukun
, Su, Junwei
, et al.
·
arXiv (Cornell University)
PDF
Reward-Aligned Reweighting for On-Policy Distillation
2026
·
Xu, Haofeng
, Su, Junwei
, et al.
·
arXiv (Cornell University)
PDF
Show more works
View as search
Works count:
4
Citation count:
9
H-index:
2
I10-index:
Year
Topic
Type
