The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 168
Lucas Castro
Castrolucasberg
·
AI & ML interests
ai safety, multimodal, model evaluation, vector databases
Recent Activity
updated a model 28 days ago
Castrolucasberg/dino-matching-lab36 published a model 28 days ago
Castrolucasberg/dino-matching-lab36Organizations
None yet
upvoted a paper 25 days ago
updated a
model 28 days ago
published a
model 28 days ago
updated a
dataset 28 days ago
published a
dataset 28 days ago