SCAR redistributes the terminal RLHF reward to tokens and spans via Shapley values, preserving the total return while improving training efficiency and final reward across three LLM alignment tasks.
Language models are few-shot learners
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.AI 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
SCAR: Shapley Credit Assignment for More Efficient RLHF
SCAR redistributes the terminal RLHF reward to tokens and spans via Shapley values, preserving the total return while improving training efficiency and final reward across three LLM alignment tasks.