SR-OPSD: Self-Referenced On-Policy Self-Distillation

• URL: http://arxiv.org/abs/2608.09745v1
• Date: Mon, 10 Aug 2026 15:40:07 GMT
• Title: SR-OPSD: Self-Referenced On-Policy Self-Distillation
• Abstract summary: On-policy self-distillation (OPSD) converts feedback into dense token-level supervision on trajectories generated by the policy to be optimized.<n>We propose the emphSelf-Referenced On-Policy Self-Distillation (SR-OPSD) to achieve state-of-the-art or competitive performance across various settings.
• Score: 18.590739659841926
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: On-policy self-distillation (OPSD) converts feedback into dense token-level supervision on trajectories generated by the policy to be optimized, providing a useful complement to reinforcement learning with sparse outcome rewards. However, the self-teacher policy used in OPSD is typically a stop-gradient or exponential-moving-average copy of the policy conditioned on additional context information, and thus co-evolves with both the student policy and its on-policy context distribution. Directly matching such a moving target with a fixed projection objective can lead to unstable optimization or excessive distributional concentration. This nature of OPSD motivates the proposed \emph{Self-Referenced On-Policy Self-Distillation (SR-OPSD)}. At fixed student-generated contexts, a token-level variational characterization identifies the effective distillation target as a geometric interpolation between the self-teacher policy and a reference policy. Meanwhile, we use the Rényi divergence family to generalize the projection geometry. This formulation separates \emph{where} the adaptive target is placed from \emph{how} the student is projected toward it: the interpolation coefficient controls underlying target, while the Rényi order controls the projection geometry and its sensitivity to token-level density ratios. Extensive experiments across scientific evaluation, mathematical reasoning, and coding generation tasks with multiple large language models show that SR-OPSD achieves the state-of-the-art or competitive performance across various settings.

Related papers

• IADD-TR: Intervention-Aware Dynamics Decoupling with Targeted Regularization for Model-Based Reinforcement Learning [57.16513298507272]
  We propose IADD-TR, a unified framework combining Intervention-Aware Dynamics Decoupling (IADD) and Targeted Regularization (TR)<n>IADD factorizes transitions into an action-intervention stage and an action-free natural evolution stage, using a zero-action anchor to resolve the non-uniqueness of this two-stage factorization for robust generalization.<n>For policy learning, we derive TR from the efficient influence function of a replay-state policy-gradient functional.
  arXiv  Detail & Related papers  (2026-08-11T08:20:02Z)
• $β$-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)
• dOPSD: On-Policy Self-Distillation for Diffusion Language Models [52.60302464420127]
  Diffusion large language models generate text by iteratively denoising a masked sequence.<n>Supervised fine-tuning is off-policy and suffers from exposure bias, while reinforcement learning gives only sparse, sequence-level rewards.<n>On-policy self-distillation (OPSD) offers a promising alternative, using one model as both student and teacher.
  arXiv  Detail & Related papers  (2026-07-05T17:47:29Z)
• Self-Distilled Policy Gradient [51.95967256628261]
  On-policy self-distillation is a promising source of dense supervision for sparse-reward reinforcement learning.<n>We propose SDPG, a self-distilled policy-gradient framework that combines group-relative verifier advantages with normalized standard deviation.
  arXiv  Detail & Related papers  (2026-06-02T02:31:13Z)
• OISD: On-Policy Internal Self-Distillation of Language Models [20.760265566511183]
  Recent reinforcement learning approaches primarily optimize the final output policy using sparse outcome-level rewards.<n>We propose the OISD framework, which improves reasoning by transferring on-policy predictive signals from the final layer to intermediate representations.<n> Experimental results demonstrate the effectiveness of OISD, with substantial and consistent improvements over strong reasoning RL baselines.
  arXiv  Detail & Related papers  (2026-05-27T20:43:10Z)
• VSPO: Vector-Steered Policy Optimization for Behavioral Control [30.80095775190934]
  We introduce Vector-Steered Policy optimization (VSPO) for language models.<n>VSPO employs a steering vector associated with the target behavior to control the behavior intensity of the generated rollouts.<n>We show that VSPO consistently improves the control along target behavior while maintaining or improving task accuracy.
  arXiv  Detail & Related papers  (2026-05-15T04:31:06Z)
• Hindsight-Anchored Policy Optimization: Turning Failure into Feedback in Sparse Reward Settings [7.554296968064274]
  We present Hindsight-Anchored Policy optimization (HAPO) for Reinforcement Learning with Verifiable Rewards (RLVR)<n>HAPO employs a hindsight mechanism that selectively anchors optimization to teacher demonstrations during failure.<n>We show that HAPO achieves textitasymptotic consistency: by naturally annealing the teacher signal as the policy improves, HAPO recovers the unbiased on-policy gradient.
  arXiv  Detail & Related papers  (2026-03-11T21:33:41Z)
• Learning Policy Representations for Steerable Behavior Synthesis [80.4542176039074]
  Given a Markov decision process (MDP), we seek to learn representations for a range of policies to facilitate behavior steering at test time.<n>We show that these representations can be approximated uniformly for a range of policies using a set-based architecture.<n>We use variational generative approach to induce a smooth latent space, and further shape it with contrastive learning so that latent distances align with differences in value functions.
  arXiv  Detail & Related papers  (2026-01-29T21:52:06Z)
• Salience-Invariant Consistent Policy Learning for Generalization in Visual Reinforcement Learning [12.9372563969007]
  Generalizing policies to unseen scenarios remains a critical challenge in visual reinforcement learning.<n>In unseen environments, distracting pixels may lead agents to extract representations containing task-irrelevant information.<n>We propose the Salience-Invariant Consistent Policy Learning algorithm, an efficient framework for zero-shot generalization.
  arXiv  Detail & Related papers  (2025-02-12T12:00:16Z)
• PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback [106.63518036538163]
  We present a novel unified bilevel optimization-based framework, textsfPARL, formulated to address the recently highlighted critical issue of policy alignment in reinforcement learning. Our framework addressed these concerns by explicitly parameterizing the distribution of the upper alignment objective (reward design) by the lower optimal variable. Our empirical results substantiate that the proposed textsfPARL can address the alignment concerns in RL by showing significant improvements.
  arXiv  Detail & Related papers  (2023-08-03T18:03:44Z)
• Implicit Distributional Reinforcement Learning [61.166030238490634]
  implicit distributional actor-critic (IDAC) built on two deep generator networks (DGNs) Semi-implicit actor (SIA) powered by a flexible policy distribution. We observe IDAC outperforms state-of-the-art algorithms on representative OpenAI Gym environments.
  arXiv  Detail & Related papers  (2020-07-13T02:52:18Z)

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.