RPO regresses the policy log-ratio against rewards centered by an empirical partition function estimated from same-prompt responses, removing the value network and reporting improved alignment, diversity, and lower toxicity.
Generalized preference optimization: A unified approach to offline alignment,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Value-Free Policy Optimization via Reward Partitioning
RPO regresses the policy log-ratio against rewards centered by an empirical partition function estimated from same-prompt responses, removing the value network and reporting improved alignment, diversity, and lower toxicity.