ODRPO decomposes discrete rewards into ordinal binary indicators to compute independent advantages and reduce noise corruption in RLAIF policy optimization.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ODRPO: Ordinal Decompositions of Discrete Rewards for Robust Policy Optimization
ODRPO decomposes discrete rewards into ordinal binary indicators to compute independent advantages and reduce noise corruption in RLAIF policy optimization.