arxiv:2603.05369
Sky
dandingsky
AI & ML interests
None yet
Recent Activity
commentedon a paper 17 days ago
Progressive Residual Warmup for Language Model Pretraining submitted a paper 18 days ago
Progressive Residual Warmup for Language Model Pretraining authored a paper 21 days ago
Thinking-Free Policy Initialization Makes Distilled Reasoning Models
More Effective and Efficient ReasonersOrganizations
models 0
None public yet
datasets 0
None public yet