Grpo
1 mentions across 0 people
All mentions
Unknown speaker
Recommendedpaper · 2026-06-10
“We apply GraphAE to GRPO, GSPO and RLOO, and conduct extensive experiments on different models and benchmarks.”
Representation-Aware Advantage Estimation Using Reward Model Hidden States for E ↗