VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training
Paper • 2602.10693 • Published • 220 •
9
floyed shen
floyed
AI & ML interests
None yet
Recent Activity
upvoted a paper 2 days ago
From Generic Correlation to Input-Specific Credit in On-Policy Self Distillation upvoted a paper 2 days ago
Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information upvoted a paper 18 days ago
Safety Instincts: LLMs Learn to Trust Their Internal Compass for Self-DefenseOrganizations
commented a
paper 2 months ago
commented a
paper 3 months ago
Upload benchmarks.zip
#2 opened 12 months ago
by 
xianghe