Skip to content

fix(grpo): use rank-penalized rewards in the advantage numerator - #646

Open
David-Wu1119 wants to merge 1 commit into
agentscope-ai:mainfrom
David-Wu1119:fix/grpo-rank-penalty-advantage
Open

David-Wu1119 wants to merge 1 commit into
agentscope-ai:mainfrom
David-Wu1119:fix/grpo-rank-penalty-advantage

fix(grpo): use rank-penalized rewards in the advantage numerator

3806e0e
Select commit
Loading
Failed to load commit list.

Workflow runs completed with no jobs