The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 549
Nathan DAVIS
bcnmarcos
·
AI & ML interests
None yet
Recent Activity
updated a model 3 days ago
bcnmarcos/mixer-multitask published a model 3 days ago
bcnmarcos/mixer-multitask updated a model 10 days ago
bcnmarcos/prompt-engineering89Organizations
None yet
upvoted a paper 10 days ago
upvoted an article 16 days ago
Article
autotrust/JEV-27B: fast, calibrated decisions and full reasoning from one open model
autotrust
• • 47