arXiv:2609.14636v1 Announce Type: new Abstract: On-policy distillation (OPD) has become a standard approach for transferring capabilities from large teachers to compact students. Its cost, however, i...
Open-MOPD addresses the capability imbalance problem in multi-teacher on-policy distillation (M-OPD) by isolating capability integration from routing ambiguity and revealing a 35.6% headroom gap compared to a domain-routed oracle ensemble. The study identifies three key factors—sequence-length disparities, convergence drift, and reward staleness—that misallocate token-level optimization budgets, leading to severe degradation in concise tasks. The proposed Open-MOPD framework introduces token-share balancing, gap-aware dynamic budget allocation, and student reward refresh, boosting headroom recovery to 83.4% and providing an open-source, reproducible post‑training recipe and evaluation suite.
arXiv:2609.36546v1 Announce Type: cross Abstract: On-policy distillation (OPD) trains a student model on its self-generated trajectories with dense token-level teacher feedback. However, naive OPD ma...
On-policy distillation (OPD) aligns a student model with a teacher's logit distribution on student-generated trajectories. This approach has achieved strong empirical gains and can often surpass conve...
arXiv:2609.37898v1 Announce Type: new Abstract: Reinforcement learning for long-horizon agents typically relies on sparse outcome-based rewards. This leads to a severe cold-start problem, as early-st...
arXiv:2609.37170v1 Announce Type: cross Abstract: Off-policy and on-policy distillation have traditionally been formulated as separate paradigms, each favoring a different property of distillation tr...
The paper investigates on‑policy distillation (OPD), showing that teacher supervision during OPD contains significant noise that grows with teacher size, yet the student policy remains largely unaffected by this noise. It finds that OPD’s gains stem mainly from suppressing low‑log‑probability tokens, a process that can be replicated without a teacher. Building on this insight, the authors propose On‑Policy Self‑Adaptation (OPSA), a supervision‑free method that uses entropy‑adaptive negative advantages to improve performance on several benchmarks, outperforming both the base model and OPD.
arXiv:2606. 21994v2 Announce Type: replace Abstract: On-policy distillation (OPD) improves reasoning models by applying dense teacher supervision on student-sampled trajectories.
The paper introduces D$^3$-MOPD, a dynamic domain scheduling method for multi-teacher on‑policy distillation. It adapts the domain mixture during training by monitoring each domain’s reverse‑KL trajectory, thereby allocating more compute to slower‑converging domains and less to those that plateau early. Experiments on a Qwen3.6‑35B‑A3B student show that D$^3$-MOPD closes 97% of the student‑to‑teacher performance gap, matches peak performance with roughly three times fewer rollout steps, and outperforms specialist teachers on most benchmarks.
arXiv:2608. 16333v1 Announce Type: cross Abstract: On-policy distillation (OPD) aligns a student model with a teacher's logit distribution on student-generated trajectories.
arXiv:2606. 24084v1 Announce Type: cross Abstract: On-policy distillation (OPD) trains a student policy using teacher signals computed on trajectories sampled by the student itself.
The paper introduces D$^3$-MOPD, a zero‑overhead scheduler that dynamically adjusts domain sampling ratios during multi‑teacher on‑policy distillation by monitoring per‑domain reverse‑KL signals. Unlike static mixtures, D$^3$-MOPD reallocates compute toward slower‑converging domains, improving efficiency and performance. In experiments with a Qwen3.6‑35B‑A3B student distilled from four domain‑expert teachers, the method closes 97% of the student‑to‑teacher gap, matches peak performance with roughly three times fewer rollout steps, and outperforms specialist teachers on most benchmarks.