REOPD: Reliability-Adaptive Reward Extrapolation for On-Policy Distillation

• URL: http://arxiv.org/abs/2608.11698v2
• Date: Thu, 13 Aug 2026 01:29:38 GMT
• Title: REOPD: Reliability-Adaptive Reward Extrapolation for On-Policy Distillation
• Abstract summary: REOPD is a reliability-adaptive reward extrapolation framework for on-policy distillation.<n>It preserves teacher alignment while selectively extrapolating along reliable teacher-reference directions.
• Score: 10.492445451268727
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: On-policy distillation (OPD) trains a student on its own trajectories under dense token-level supervision from a teacher. Reward-extrapolation methods such as ExOPD amplify the teacher-reference log-likelihood ratio to move beyond direct imitation, but apply a single global coefficient $λ$ to every token. This can drive the student to fit extreme peaks in the implicit reward, causing reward hacking and unstable training, and the optimal $λ$ varies across domains, requiring costly sweeps. We propose REOPD, a reliability-adaptive reward extrapolation framework for OPD. REOPD combines a token-level compatibility weight with a batch-level adaptive budget, yielding a token-wise coefficient $λ_{b,t}=1+γ_b q_t$ that preserves teacher alignment while selectively extrapolating along reliable teacher-reference directions. It requires no verifier, reward model, value model, or extra rollout beyond standard OPD. REOPD outperforms G-OPD on single-teacher mathematics and on both domains in the multi-teacher setting, while matching G-OPD on single-teacher code, demonstrating effective fine-grained reliability adaptation across domains and teacher configurations.

Related papers

• Self-OPD: On-Policy Distillation for Flow Matching Models without Teacher [36.07614733558147]
  We introduce textbfSelf-OPD, a teacher-free OPD framework for flow matching models.<n>Self-OPD branches the deterministic next-state prediction into $K$ SDE candidates, rolls them out with the ODE sampler, and compares their rewards against a deterministic self-reference baseline.<n>Experiments on single and mixed reward benchmarks show that Self-OPD outperforms prior RL and OPD methods without task-specific teachers.
  arXiv  Detail & Related papers  (2026-08-27T09:34:23Z)
• ReOrder-OPD:Reliability-Aware Prompt Ordering for On-Policy Distillation [25.69241346799495]
  We define prompt-level teacher continuation reliability $R$ as the teacher's probability of reaching a correct answer from a student prefix.<n>We show that high-$R$ prompts yield larger OPD gains and that descending-$R$ training outperforms random and ascending orders.
  arXiv  Detail & Related papers  (2026-08-11T13:27:56Z)
• Simple-OPD: Demystifying Warm-up for On-policy Distillation [51.693357695326846]
  On-policy distillation (OPD) trains a student on its own rollouts with token-level supervision from teacher models.<n>We demystify warm-up for OPD from both data and training perspectives.
  arXiv  Detail & Related papers  (2026-08-07T04:47:38Z)
• Distill Where You Fail: Recovering Learning Signals of Negative RL-Groups from Adaptive Teacher Guidance [50.77757355236912]
  We propose RSTG (Recovering Learning Signals via Adaptive Teacher Guidance), which applies distillation selectively and precisely where it matters most.<n> Experiments demonstrate that RSTG substantially outperforms naive GRPO+OPD by +4.02% on math and+3.05% on code.
  arXiv  Detail & Related papers  (2026-08-01T17:29:14Z)
• $β$-OPSD: Deriving with Policy Optimization, Training with Self-Distillation [85.07190808882523]
  On-policy self-distillation is a promising approach to improve reasoning language models.<n>We show that $$-OPSD consistently outperforms vanilla OPSD.
  arXiv  Detail & Related papers  (2026-07-30T17:41:16Z)
• H$^2$SD: Hybrid Hindsight Self-Distillation [61.71131241473028]
  Reinforcement learning with verifiable rewards (RLVR) provides reliable outcome supervision for language model reasoning.<n>Existing self-distillation methods add a privileged teacher but typically assign it a fixed role.<n>We introduce Hybrid Hindsight Self-Distillation ($mathrmH2mathrmSD$), which jointly adapts teacher context and update strategy to trajectory correctness.
  arXiv  Detail & Related papers  (2026-07-21T10:47:27Z)
• SAGE-OPD: Selective Agent-Guided Intervention for Multi-Turn On-Policy Distillation [25.25989941933095]
  We propose SAGE-OPD, a verifier-free selective intervention framework specifically designed for multi-turn OPD.<n>Instead of applying teacher supervision uniformly across all turns, SAGE-OPD first observes environment feedback and uses teacher judgment to decide whether each student response should be skipped or intervened on.<n>Experiments on agent tasks show that SAGE-OPD consistently improves over baselines, achieving up to a 13.3% relative improvement in ALFWorld unseen success rate over standard OPD.
  arXiv  Detail & Related papers  (2026-06-17T23:58:14Z)
• SG-OPD: Sign-Gated On-Policy Distillation via Sign-Consistency Gating and Phased Teacher Sampling [22.832360652920332]
  On-policy distillation (OPD) trains a student on its own trajectories with dense per-token supervision from a stronger teacher.<n>We propose Sign-Gated On-Policy Distillation (SG-OPD), which uses a binary verifier as a trust signal for the teacher.<n> Experiments on competition-level mathematical reasoning benchmarks show that SG-OPD consistently outperforms standard OPD.
  arXiv  Detail & Related papers  (2026-06-08T10:11:58Z)
• On-Policy Distillation with Best-of-N Teacher Rollout Selection [54.91780727674628]
  We propose BRTS, a Best-of-N Rollout Teacher Selection framework for on-policy distillation.<n>BRTS augments standard student-context OPD with a teacher-context supervision branch constructed from the teacher trajectory.<n>BRTS improves over standard OPD on challenging reasoning benchmarks, with the largest gains on harder datasets.
  arXiv  Detail & Related papers  (2026-05-10T19:49:00Z)
• Prune-OPD: Efficient and Reliable On-Policy Distillation for Long-Horizon Reasoning [66.52232008796294]
  Prune-OPD dynamically aligns training budgets with supervision quality.<n>It reduces training time by 37.6%--68.0% while preserving, and often improving, performance on challenging benchmarks.
  arXiv  Detail & Related papers  (2026-05-08T14:38:53Z)
• Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation [57.524909883706556]
  On-policy distillation (OPD) has demonstrated strong empirical gains in improving student performance.<n>This work introduces a flexible reference model and a reward scaling factor that controls the relative weight of the reward term against the KL regularization.<n>In particular, in the setting where we merge the knowledge from different domain experts, obtained by applying domain-specific RL to the same student model, ExOPD enables the student to even surpass the teacher's performance boundary.
  arXiv  Detail & Related papers  (2026-02-12T16:14:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.