arxiv:2410.04717
Dylan PRO
shizhuo2
AI & ML interests
None yet
Recent Activity
upvoted a paper 3 days ago
Selecting Diverse SFT Traces Improves Post-RL Generalization submitted a paper 3 days ago
Selecting Diverse SFT Traces Improves Post-RL Generalization upvoted a paper 3 days ago
Learning from Teacher Continuations at Student States