The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 208
विक्रम शर्मा
Kabirvkv
·
AI & ML interests
Into NLP, diffusion models and AI agents.
Recent Activity
upvoted a paper 30 days ago
Training Object Permanence in World Models updated a model about 1 month ago
Kabirvkv/dissertation-multitaskOrganizations
None yet