Explore
Discover
Solutions
AI Services
Our Products
Submit a tool
Search the catalog
⌘K
$
/
₹
Sign up free
Login
Back to papers
November 10, 2026
cs.LG
stat.ML
A Theory of Policy Optimization in Post-Training: Coverage Collapse and Reference Resets
Audrey Huang
,
Nan Jiang
,
Sham M. Kakade
,
Sadhika Malladi
,
Akshay Krishnamurthy
,
Dylan J. Foster
Original Abstract
Read on arXiv
Download PDF
Categories
cs.LG, stat.ML
A Theory of Policy Optimization in Post-Training: Coverage Collapse and Reference Resets - AI | One9Founders