The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 168
Chaemin Nam
chaeminnam
AI & ML interests
None yet
Recent Activity
upvoted a paper 29 days ago
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search AgentsOrganizations
None yet
When constraints cannot be satisfied, does the system identify the conflicting pair?
1
#292 opened 1 day ago
by chaeminnam