ReNIO: Reweighting Negative Trajectory Importance for LLM On-Policy Distillation
Paper • 2606.23104 • Published • 3
Chen Lin
Alephia
·
AI & ML interests
LLM, Knowledge Distillation, Post Training
Recent Activity
updated a model almost 2 years ago
Alephia/tryOrganizations
None yet
upvoted a paper 1 day ago