The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 208
Clara Schroeder
SCHROEDERva
·
AI & ML interests
Interested in machine learning, NLP and open source AI.
Recent Activity
upvoted a paper 19 days ago
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents upvoted a paper 19 days ago
Does Learning Protein Folding Generalize to Broader Reasoning?Organizations
None yet
upvoted 3
papers 19 days ago