FiMi-RM reduces length bias in RLHF reward models by fitting a nonlinear length-reward curve and training the reward model to be uncorrelated with that fitted curve.
Noise contrastive alignment of language models with explicit rewards, 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
other 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
other 1polarities
unclear 1representative citing papers
citing papers explorer
-
Bias Fitting to Mitigate Length Bias of Reward Model in RLHF
FiMi-RM reduces length bias in RLHF reward models by fitting a nonlinear length-reward curve and training the reward model to be uncorrelated with that fitted curve.