The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 418
Zihao HU
Hu766144
·
AI & ML interests
Transformer enthusiast since 2019
Recent Activity
upvoted a paper 2 days ago
VisionHOPE: Visual Backbones as Self-Modifying Learning Systems upvoted a paper 2 days ago
In-Context Learning for Robots: Methods and ApplicationsOrganizations
None yet
upvoted 3
papers 2 days ago