Skip to content

feat: enhance advantages tracking and normalization stability in GRPO#1423

Merged
terrykong merged 5 commits intomainfrom
ffrujeri/grpo_improvements
Nov 13, 2025
Merged

feat: enhance advantages tracking and normalization stability in GRPO#1423
terrykong merged 5 commits intomainfrom
ffrujeri/grpo_improvements

Commits

Commits on Nov 11, 2025