Eliciting Weak-to-Strong Generalization with On-Policy Reverse Distillation

• URL: http://arxiv.org/abs/2609.08798v1
• Date: Tue, 08 Sep 2026 14:26:35 GMT
• Title: Eliciting Weak-to-Strong Generalization with On-Policy Reverse Distillation
• Abstract summary: Weak-to-strong generalization asks whether stronger models can learn from weaker supervisors and surpass them.<n>We introduce On-Policy Reverse Distillation, which evaluates the teacher's policy shift and amplifies the student's verifier-driven policy gradient.<n>In both successive model transfer and multi-teacher distillation, OPRD achieves prohibitively higher performance with fewer student updates than existing RL and distillation approaches.
• Score: 41.854144563456046
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Weak-to-strong generalization asks whether stronger models can learn from weaker supervisors and surpass them. This question is particularly important for successive model generations and multi-domain consolidation, where repeating frontier-scale post-training from scratch can be prohibitively expensive. Yet conventional distillation treats the weak teacher as an optimization target, potentially imposing its capacity ceiling on the student. We introduce On-Policy Reverse Distillation (OPRD), which evaluates the teacher's policy shift relative to its reference policy on student rollouts and amplifies the component of the student's verifier-driven policy gradient along that direction. By rescaling only verifier-supported updates, OPRD preserves the stationary points of policy optimization while accelerating learning beyond the teacher. In both successive model transfer and multi-teacher distillation, OPRD achieves higher performance with fewer student updates than existing RL and distillation approaches. Response-style analysis shows that OPRD students remain closer to models trained with verifier-based RL alone than to their weak teachers, suggesting that teacher guidance accelerates rather than redirects the student's own optimization. Results in conventional strong-to-weak distillation further demonstrate that OPRD effectively combines verifier-driven policy optimization with teacher guidance regardless of capacity ordering.

Related papers

• 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)
• Teaching the Way, Not the Answer: Privileged Tutoring Distillation for Multimodal Policy Optimization [22.479229036646156]
  We propose a Privileged Tutoring Distillation Policy Optimization framework for Reinforcement Learning with Verifiable Rewards (RLVR)<n> PTD-PO provides dense guidance without exposing the answer to the student policy.<n> Experiments on LVLMs ranging from 2B to 8B parameters show that PTD-PO consistently outperforms RLVR and distillation baselines.
  arXiv  Detail & Related papers  (2026-06-05T07:43:22Z)
• SocraticPO: Policy Optimization via Interactive Guidance [64.7524628699057]
  Reinforcement learning (RL) for large language models usually supervises reasoning with scalar outcome rewards, such as binary correctness.<n>We propose textbfSocraticPO (Socratic Policy Optimization), a policy-optimization framework that augments RL rollouts with Socratic-style natural-language guidance.
  arXiv  Detail & Related papers  (2026-06-03T09:08:29Z)
• Prefix Teach, Suffix Fade: Local Teachability Collapse in Strong-to-Weak On-Policy Distillation [49.117085054884676]
  On-policy distillation trains a student model on its own rollouts using dense feedback from a stronger teacher.<n>We operationalize this principle through a trajectory-specific release rule.<n> Experimental results across strong-to-weak distillation tasks indicate that this release rule consistently outperforms standard full-trajectory OPD.
  arXiv  Detail & Related papers  (2026-05-13T15:05:30Z)
• Teacher-Guided Policy Optimization for LLM Distillation [90.49982387646861]
  Teacher-Guided Policy Optimization (TGPO) is an on-policy algorithm that incorporates dense directional guidance by leveraging teacher predictions conditioned on the student's rollout.<n>Experiments on complex reasoning benchmarks demonstrate that TGPO significantly outperforms standard baselines and is robust to different teachers.
  arXiv  Detail & Related papers  (2026-05-13T09:20:03Z)
• Preference-Based Self-Distillation: Beyond KL Matching via Reward Regularization [18.027254451537342]
  Existing self-distillation methods largely reduce learning to KL matching toward the context-augmented teacher model.<n>We propose textbfPreference-textbfBased textbfSelf-textbfDistillation (textbfPBSD), which revisits on-policy self-distillation through a reward-regularized perspective.
  arXiv  Detail & Related papers  (2026-05-06T15:31:50Z)
• Reinforcement-aware Knowledge Distillation for LLM Reasoning [63.53679456364683]
  Reinforcement learning (RL) post-training has recently driven gains in long chain-of-thought reasoning large language models (LLMs)<n>Most existing knowledge distillation methods are designed for supervised fine-tuning (SFT), relying on fixed teacher traces or teacher-student Kullback-Leibler (KL) divergence-based regularization.<n>We propose RL-aware distillation (RLAD), which performs selective imitation during RL -- guiding the student toward the teacher only when it improves the current policy update.
  arXiv  Detail & Related papers  (2026-02-26T00:20:39Z)
• 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)
• KEPO: Knowledge-Enhanced Preference Optimization for Reinforcement Learning with Reasoning [24.072603982041798]
  Reinforcement learning has emerged as a promising paradigm for inducing explicit reasoning behaviors in large language and vision-language models.<n>However, reasoning-oriented RL post-training remains fundamentally challenging due to sparse trajectory-level rewards.<n>Recent on-policy distillation methods introduce dense teacher supervision to stabilize optimization, but apply it uniformly across all generated trajectories.
  arXiv  Detail & Related papers  (2026-01-30T23:28:37Z)

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.