This work proposes RetireOPD (Self-Retiring On-Policy Distillation), which first optimizes a decoupled, skill-conditioned teacher with environment rewards and then trains a skill-free student jointly with RL and OPD.
Yan Yu, Zhengxi Lu, Yi-Zhou Liu et al.· 4 citations
SERTPO (Self-Evolving Rubric Policy Optimization), which replaces answer voting with a closed loop that co-evolves response evidence, query-specific rubrics, and policy parameters, and supports OOD transfer and continued cross-benchmark evolution.
Jian-Ze Wang, Kun-Wang Zheng, Ying Liu et al.· arXiv.org· 2 citations
Test-Time Policy Optimization is proposed, an asymmetric objective that distills agreeing rollouts via OPSD and penalizes disagreeing rollouts with Grouped RL and Token-level selection further refines both branches: distillation down-weights already-converged positions, while RL penalizes only confident errors.
Ao-Han Wang, Zhengxi Lu, Jian-Ze Wang et al.· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.