VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training
Paper
• 2602.10693 • Published
• 167
Xiang Cheng
FFFc2
·
AI & ML interests
None yet
Recent Activity
upvoted a collection about 12 hours ago
dots.ocr authored
a paper
7 days ago
VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training Organizations
None yet
authored a paper 7 days ago