Sage Self Hinting Rl

Inject privileged hints during GRPO training to reshape rollout distributions when advantage collapses occur, increasing outcome diversity without changing task rewards. Hints removed at deployment; policy automatically learns when to use hints via online refresh mechanism.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit 11c57f926d

Frequently asked questions

npx get-skill add adu2021/sage-self-hinting-rl