RED assigns token-level rewards by taking the difference of a sequence reward model's scores on adjacent prefixes, improving RLHF training without additional reward-model training.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
RED: Unleashing Token-Level Rewards from Holistic Feedback via Reward Redistribution
RED assigns token-level rewards by taking the difference of a sequence reward model's scores on adjacent prefixes, improving RLHF training without additional reward-model training.