SHAPE improves average math reasoning accuracy by 3% while cutting token use by 30% through stage-aware hierarchical advantage and entropy-driven token redistribution.
Substituting these conditions into the reward equa- tion: min τ + Rtotal(τ +) = 1+α(γ min·0−1) = 1−α(13)
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
-
SHAPE: Stage-aware Hierarchical Advantage via Potential Estimation for LLM Reasoning
SHAPE improves average math reasoning accuracy by 3% while cutting token use by 30% through stage-aware hierarchical advantage and entropy-driven token redistribution.