Defa Zhu

mathfinder

·

https://zhudefa.github.io/

mathfinder

AI & ML interests

None yet

Recent Activity

upvoted a paper 2 days ago

Hierarchical Residuals: A Minimalist Decoupling of Residual Width from Compute Width

upvoted a paper 9 days ago

Your LR Schedule May Not Be Principled: Effective Learning Rate Scheduling in Weight-Normalized Training

upvoted a paper 15 days ago

Rollback: Experience Backtracking in World Models for Policy Improvement

View all activity

Organizations

None yet

mathfinder 's papers 8

arxiv:2511.11238

arxiv:2508.18756

arxiv:2504.13914

arxiv:2503.16057

arxiv:2503.14125

arxiv:2501.16975

arxiv:2411.12364

arxiv:2409.19606