Rationale-Guided Policy Optimization: Learning to Reason with Adaptive Rationale Scaffolding
Open paper
Benchmark
Model
Rank
Results
mathematical-reasoning-on-minerva-math
RGPO
#41
Accuracy: 22.1
mathematical-reasoning-on-olympiadbench
RGPO
#48
Accuracy: 24.9
mathematical-reasoning-on-math-500
RGPO
#121
Accuracy: 55.2
Rank counts only results with a code link.