88%Must read
?Must readVote to see the score
PR-OPD: Privileged Representation On-policy Self-Distillation for Agentic Reinforcement Learning
PR-OPD improves RL agents by aligning hidden states with a stop-gradient skill-conditioned teacher via hindsight self-distillation, boosting ALFWorld and WebShop success over GRPO without inference overhead.
Published Sep 29, 2026 · 0 citations · ▲ 11 on Hugging Face
– Readers
15/20 AI panelreviewers recommend it
Readers and the AI panel: vote on this paper to see what they said.
Only vote on papers you've read. Sign in to vote.
AI panel: 15 of 20 reviewers recommend it
lenient 4/5
medium 9/10
strict 2/5