Overview

On-policy distillation studies how to transfer a teacher’s capabilities by training a student on trajectories the student generates, rather than only on fixed teacher-written examples. Common approaches match teacher and student token distributions with KL-style objectives, often combining this dense feedback with verifier rewards, reinforcement learning, or feedback from successful peer attempts. Work spans language-model reasoning and tool-using agents as well as multimodal, speech, diffusion, and embodied models; methods increasingly select useful tokens, turns, or rollout lengths to reduce compute. A central tension is that student-generated states can make supervision more relevant, but teacher signals may be unreliable or mismatched, and repeated rollouts can be costly or destabilize training. 4 sentences total.

Latest papers

551 papers88 papers in the last four weeks, down 49% on the four weeks before. 0.6% of all new papers.

Nov 12, 2026

Teach with Blinks: Breaking Linguistic Momentum in On-Policy Distillation for Fine-Grained Multimodal Perception

On-policy distillation (OPD) has emerged as an effective post-training approach to improving fine-grained perception in multimodal large language models by amortizing costly inference-time zooming operations into a single-pass recipe. However, vanilla OPD is inherently susceptible to linguistic momentum: the teacher can follow the student's erroneous prefixes, thus weakening corrective supervision. To address this, we propose Blink On-Policy Distillation (BlinkOPD), in which the teacher "blinks" at every token: it reads the same prefix once with its eyes open and once with its eyes closed. A first-glance gain further strengthens the correction on rollouts that conflict with the teacher. By extracting more effective supervision from each on-policy trajectory, BlinkOPD achieves state-of-the-art average performance across six fine-grained benchmarks using only 500 training images (12x less data) than existing OPD methods. Beyond fine-grained perception, BlinkOPD also demonstrates strong zero-shot improvements across ten held-out benchmarks spanning general visual perception, visual hallucination, and multimodal reasoning abilities. Code: https://github.com/Supltz/BlinkOPD
Nov 12, 2026

Turning Multi-Agent Communication into Token-Level Credit with On Policy Self Distillation

Reinforcement learning improves problem solving in large language model multi-agent systems. Existing approaches overlook communication knowledge or compress it into scalar rewards, leaving coarse credit with little guidance on which decisions each agent should reinforce or revise. OPSD4MAS converts future communication into fine-grained token-level credit through On Policy Self Distillation (OPSD). An external model extracts facts and assesses their trajectory relations for coarse-grained filtering, followed by fine-grained leave-one-out filtering of target-token influence against agent-wise outcome advantages. The fact-conditioned policy supplies token-level credit, combined with GRPO through a linearly annealed joint objective. Our method improves mathematical pass@4 over GRPO by up to 20.00 percentage points. Ablations support complete filtering, while frozen-history injection diagnostics support fact utility. Together, these results support combining agent-level outcome guidance with communication-derived token-level credit. Code is available at https://github.com/Camel-Prince/OPSD4MAS.
Nov 11, 2026

Janus-MOPD: Tracing the Student Learning Dynamics in Multi-Teacher On-Policy Distillation

Multi-domain capabilities are critical for practically applicable Large Reasoning Models (LRMs). Therefore, Multi-teacher On-Policy Distillation (MOPD) has become a crucial part of the post-training process, integrating the capabilities of multiple domain-specific teacher models into a single student model. However, the student usually fails to acquire multi-domain capabilities simultaneously due to undesirable domain heterogeneity. Existing works have address this problem in view of temporal uneven efficiency and spatial hard allocation in external multi-teacher supervision. While in broader multi-domain training, varying degrees of performance degradation is still largely unexplained. In this paper, we shift the focus on how the student learns and integrates multi-domain capabilities internally, and then propose Janus-MOPD, a general learning dynamics framework of geometric representations for LRMs. Specifically, our main focus is on the internal performance of multi-domain capabilities on both temporal and spatial dimensions within the student. Temporally, our study shows that the student exhibits a Expansion-Consolidation-Optimization (ECO) three-phase learning dynamics, together with inherent limitations in learning efficiency and scaling. Spatially, we summarize two types of subspace locking within and across domains by tracing the representation space of the student, while explaining its clear learning directions but inherent cross-domain conflict. Therefore, Janus-MOPD yields a unified learning dynamics formulation, a geometric tracing methodology, and systematic findings of how the student learns and integrates multi-domain capabilities. Overall, we provide a Janus perspective to internally trace the continual evolution of the multi-domain capabilities within the model itself.
Nov 10, 2026

Coding Agent Memory Post-Training II: Unlocking the Memory Potential of Pre-trained File Operations for Long-Horizon Tasks via On-Policy Distillation

Language-model agents increasingly tackle long-horizon tasks whose interaction histories exceed the model's active context. They must therefore decide what to write down and what to read back, and when to do either. Recent work has begun to use on-policy distillation (OPD) to make memory control part of the policy, relying on memory tools and compaction protocols that lie outside either model's pre-training. However, this weakens the distillation on both sides: the teacher cannot give useful guidance on an interface it has never used, and the student samples too few meaningful memory actions for that guidance to take effect. To study this problem, we introduce Coding Agent Memory Gym II (CAMG-II), a suite of long-horizon agentic-RL environments spanning Shop, Coding, AutoResearch, General, and Visual. Alongside each environment's native task interface, CAMG-II provides ordinary Bash access and an episode-persistent workspace, enabling the policy to create, revise, search, and reuse files as memory across context replacements. We also introduce Coding Agent Memory Post-Training II (CAMP-II), which trains a single policy jointly across all five environments with fully asynchronous OPD to learn this file-based memory behavior directly from a frozen teacher with a prior for file operations. On CAMG-II, CAMP-II obtains 2.97 times the gain of the strongest OPD baseline. The memory behavior CAMP-II learns generalizes to two external benchmarks, Terminal-Bench 2.1 and SWE-Bench Pro V2, with improved success rates.
Nov 10, 2026

Beyond Passive Imitation: On-Policy Mid-Training Distillation for Language Models

Continual pre-training (CPT) adapts large language models (LLMs) to specialized domains, yet its standard next-token prediction (NTP) objective treats every corpus token as a fixed target, regardless of its utility to the current model. In a controlled study on the same 1B-token mathematical corpus, we observe sharply different outcomes across base models: NTP improves Llama-3.2-3B's performance on mathematical benchmarks but substantially degrades Qwen3-1.7B-Base's performance, indicating that, under a fixed training recipe, the same training sequences can have model-dependent effects. To address this problem, we introduce On-Policy Mid-Training Distillation (OPMD), which reformulates mid-training from offline text imitation into active, query-driven knowledge acquisition. At selected anchor positions, OPMD samples trajectories from the evolving student policy and applies dense token-level reverse-KL rewards from a frozen teacher, training the model on states it actually visits rather than only on ground-truth prefixes. This on-policy objective reduces the training-generation mismatch of teacher-forced NTP. On MATH-500 and Minerva, OPMD improves the Qwen student over its base performance and outperforms the evaluated NTP baseline, which degrades on both benchmarks. OPMD integrates into CPT pipelines without added inference-time overhead and offers a model-aware alternative to passive domain adaptation.
Nov 10, 2026

Bootstrapped On-Policy Distillation

On-policy distillation (OPD) improves language models by distilling teacher predictions on student-generated trajectories, while on-policy self-distillation (OPSD) removes the need for an external teacher by using a reference-conditioned copy of the model as the teacher. Despite its great early improvements, we find that OPSD does not remain consistently effective under continued training and can eventually suffer substantial performance degradation. We study this training behavior and link the degradation to a growing mismatch between the evolving reference-free student and the frozen privileged teacher. In the early stage of OPSD, the privileged teacher remains closely aligned with the student and provides useful guidance for improving the student's generation trajectories. However, as training progresses, the growing distributional mismatch between the privileged teacher and reference-free student can cause the teacher to provide misleading supervision, ultimately degrading student performance. This phenomenon motivates us to first leverage privileged reference supervision to acquire useful knowledge and then transfer it to a fresh student. We introduce Bootstrapped On-Policy Distillation (BOPD), a two-stage framework that first uses OPSD to internalize task-relevant information from reference-conditioned supervision into an intermediate model, and then distills this reference-free intermediate model into a fresh copy of the initial model. Experiments across models show that BOPD improves mathematical reasoning and code generation beyond both the initial models and the constructed OPSD teachers. These results suggest that privileged reference supervision can be more effectively exploited by first transferring its task-relevant signal into a task-adapted model and then distilling from that model without reference conditioning.
Nov 9, 2026

Domain-Routed Delta Optimization for Multi-Teacher On-Policy Distillation

Multi-teacher on-policy distillation can yield uneven domain capabilities. We propose Domain-Routed Delta Optimization (DRD-MOPD), composing a frozen base with expert-initialized trainable deltas. Domains update their deltas with independent optimizer states and clipping. After joint training, Post-separate exports the base plus one trained delta per domain, selected by known task domain without retraining. Using AdamW, the unweighted full student reaches 32.83% Macro across six benchmarks after 20 updates, versus 32.21% for a matched shared control. Post-separate raises Macro with Open-MOPD weighting.
Nov 9, 2026

Which Successes Make Good References? Action-Sequence Alignment for Multi-Turn On-Policy Distillation

On-policy distillation (OPD) reduces train-inference distribution mismatch by supervising a student on its own trajectories. For multi-turn tool-use agents, conditioning the teacher on a validated same-task successful trajectory as a reference can provide useful hindsight, but different successes may reach incompatible interaction states and offer misleading supervision. We cast successful-reference selection as a free-target optimal transport problem, which reduces to assigning each failed trajectory its most compatible successful reference. Because latent state paths are hidden, we use action-sequence alignment as an observable proxy for compatibility and instantiate it with longest common subsequence (LCS), which identifies a shared ordered backbone of actions while allowing recoverable detours. This perspective makes trajectory compatibility the central criterion for reference selection in multi-turn OPD. Experiments on ALFWorld and the airline domain of τ3τ^3-bench show that LCS is the only evaluated reference selector to improve over reference-free OPD on both ALFWorld overall scores and all three airline metrics. Without further optimization, the airline-trained models also transfer to the retail domain, where LCS improves all three metrics.
Nov 9, 2026

Agentic-OPD: Recovering Forgotten Agent Capabilities via On-Policy Distillation for Agentic Tasks

Reinforcement learning is now the mainstream paradigm for training LLMs on agentic tasks, yet standard practice keeps only the best-validation checkpoint and discards the rest. We show that this is lossy: in GRPO training of LLM agents, the best-validation checkpoint fails on problems that earlier checkpoints from the same run had already solved, and the union of problems solved across checkpoints far exceeds any single one. The discarded checkpoints thus form a teacher pool obtained at no additional training cost, which on-policy distillation (OPD) can mine to recover the lost coverage. Mining this pool raises two challenges: i) which checkpoint to distill from, as no checkpoint dominates the others and complementarity to the student is problem-specific; ii) where to trust it, as even a well-chosen teacher is only locally superior, so indiscriminate distillation overwrites behavior the student has already mastered. We propose Agentic-OPD, which treats the best-validation checkpoint as the student and all saved checkpoints as frozen teacher candidates, using the student's evolving capability gaps to guide both teacher selection and supervision filtering: the HCN policy (Highest Coverage Next) dynamically selects teachers by weighted coverage of the student's unsolved problems, while the LAD module (Locality-Aware Distillation) confines transfer to failed student rollouts on problems where the teacher performs better. Across five agentic task domains and 24 benchmarks and task subsets, Agentic-OPD improves over its initialization checkpoint by up to 7.4 points, with a 4B student surpassing the latest frontier models GPT-5.6 and DeepSeek-V4.1 on deep search. Agentic-OPD recasts a finished RL run as a reusable pool of policies and establishes post-RL capability recovery as a stage orthogonal to improvements in the RL algorithm itself.
Nov 9, 2026

VLA-MOPD: On-Policy Distillation for Multi-Task VLA Policy Consolidation

Physical intelligence requires embodied agents to ground language in observations and execute actions while retaining useful skills. Vision-language-action (VLA) policies offer a shared interface, but task-specific post-training often produces separate specialists. Consolidation is difficult: mixed-task reinforcement learning provides sparse trajectory-level credit, while offline distillation may miss student-visited observations and endpoint distillation does not supervise intermediate action-generation states. We introduce VLA Multi-teacher On-Policy Distillation (VLA-MOPD), a protocol for consolidating flow-matching VLA specialists. After acquiring task or suite experts, a shared student collects robot rollouts and queries the instruction-routed frozen teacher at each visited observation and intermediate ordinary differential equation (ODE) state. Detached per-flow-step transition targets supervise valid action chunks through a masked local matching loss; an optional reference-policy term supports retention on other tasks. Starting from one trajectory per LIBERO task, VLA-MOPD raises mean success-once from 75.4% to 98.9%, compared with 98.15% for endpoint action distillation. It improves mean RoboTwin 2.0 success from 73.0% to 89.3%; on a Piper robot, two consolidation rounds improve average success from 70.0% to 90.0%. In a near-matched Spatial-adaptation study, reference matching limits the Object suite decrease to 3.0 points, versus 13.2 without the reference term.
Nov 9, 2026

ECOS: Environment-Calibrated On-Policy Self-Distillation for Language Agents

On-policy self-distillation lets a language agent act as its own Teacher by conditioning on privileged context such as skills written from its own trajectories. Its guidance, however, is weighted by how much this context shifts token likelihoods rather than by whether it improves behavior, so errors in self-written skills return through later updates. We propose environment-calibrated on-policy self-distillation, which uses environmental outcomes to decide how much to trust the Teacher's current guidance and which self-written skills to reinforce. Command-level advantages, estimated by comparing commands executed at similar observations, scale each Teacher preference and reward reflections and skills whose induced preferences favor higher-advantage commands. Without skills at inference, 3B and 7B agents reach 90.5% and 93.3% macro success on seen ALFWorld tasks and 75.5% and 80.5% success on WebShop. On unseen ALFWorld tasks, they outperform GRPO under the same protocol by 11.0 and 7.6 percentage points, respectively. In an ablation study, supervising only executable commands rather than the Actor's full response raises 3B WebShop success from 52.3% to 75.0%.
Nov 9, 2026

Robust Domain-Mixture Scheduling for Multi-Teacher On-Policy Distillation

Multi-teacher on-policy distillation (MOPD) merges the capabilities of multiple domain-expert teachers, such as mathematical reasoning, code generation, and instruction following, into a single compact student. Since all domains share a limited training budget, how to allocate it across domains directly determines the quality of distillation. Existing methods allocate the budget by each domain's learning progress, but offer no guarantee for lagging domains. We therefore introduce distributionally robust optimization (DRO) into MOPD to prioritize the worst-performing domain. However, the per-domain training signals are estimated from a few rollouts and are inherently noisy, so directly applying DRO mistakes noise for the worst case, lets occasional spikes hijack the budget, and protects inherently hard domains rather than truly regressing ones. To address these issues, we propose RDS-MOPD, which keeps the DRO objective and redesigns its three parts: it shifts the allocation only when differences clearly exceed the noise, bounds the influence of any single anomaly, and defines the worst case as the regression from each domain's own best level, all with theoretical guarantees. Experiments show that RDS-MOPD outperforms recent competing methods and ranks first on most mathematical reasoning, code generation, and instruction-following benchmarks.
Nov 8, 2026

How Much Training-Inference Mismatch Can On-Policy Distillation Tolerate?

On-policy distillation (OPD) trains a student language model on its own responses, guided by a fixed teacher. Generating these responses is expensive, so systems use quantized inference and asynchronous rollouts to reduce the cost. Both, however, introduce training-inference mismatch between the rollout engine and the model being trained. Such mismatch can destabilize reinforcement learning (RL), but its effects on OPD remain unclear. This paper investigates how much mismatch OPD tolerates, where its effects arise, and how training systems can use this tolerance. Measuring all mismatch sources on one scale, we examine quantization across three student-teacher pairs, with additional tests of weight staleness and injected noise on one pair. OPD tolerates substantial mismatch without meaningful loss of training quality but degrades at higher levels, revealing an empirical mismatch budget. Both the quality cost of mismatch and the mismatch introduced by quantization depend on the model, with some layers particularly sensitive to reduced precision. In controlled comparisons, OPD remains stable under rollout configurations that destabilize RL methods. Under weight noise, mismatch also shifts the prefixes reached during generation, introducing errors that per-token corrections cannot address. To put these findings into practice, we introduce a lightweight admission check that screens rollout configurations against a calibrated budget without requiring a full training run for each candidate. Model-specific FP8 recipes preserve training quality while providing capacity for at least 1.93×1.93\times as many KV-cache tokens as BF16. Combining FP8 with asynchronous rollouts reaches the same training quality 1.62×1.62\times sooner than synchronous BF16 training. Code is available at https://github.com/benyucong/train-inference-mismatch-opd
Nov 5, 2026

Unpacking the Compute Efficiency of Strong-to-Weak OPD against GRPO

On-policy distillation (OPD) trains a student on its own rollouts under a frozen teacher's per-token supervision. Does this dense signal reduce the training FLOPs needed to reach a target accuracy compared with reinforcement learning? We systematically compare vanilla reverse-KL OPD with Group Relative Policy Optimization (GRPO) across task difficulty, student and teacher size, domain, and optimization choices, measuring held-out accuracy against training FLOPs. With an external post-trained teacher, OPD generally loses to GRPO in accuracy per training FLOP at matched settings, even when the teacher is further trained with GRPO, and often suffers response-length inflation. Yet a different teacher changes the picture: when the teacher is the student's own GRPO checkpoint, OPD recovers that checkpoint's accuracy on math with 50×50\times fewer distillation FLOPs than GRPO needs to reach the same accuracy. Following the same teacher setting, we consistently observe this high efficiency with a larger student, on code, and in a second model family. On a single domain, however, the GRPO checkpoint is already the desired model, so distilling it adds to total compute. Across domains, separate GRPO checkpoints hold different capabilities, and no single checkpoint serves every domain. Multi-teacher OPD (DeepSeek-AI, 2026) consolidates per-domain GRPO experts into a single student that matches each expert's accuracy, with the distillation stage costing only 5.2% (on 2 domains) and 8.0% (on 4 domains) of the specialist training FLOPs. Counting specialist training and distillation together, this pipeline reaches joint GRPO's performance on the 2-domain mix with 42.0% fewer total FLOPs.
Nov 5, 2026

Learning Dynamics of On-Policy Distillation

On-policy distillation (OPD) is emerging as an important paradigm for language-model post-training, combining dense teacher supervision with training on student-generated trajectories. This creates a feedback loop where each update changes the student and thereby the training distribution that generates subsequent updates. Yet, how this feedback shapes learning and generalization remains poorly understood. We conduct a mechanistic analysis of OPD's learning dynamics, distinguishing the two feedback loop stages: conversion of the distillation signal into parameter updates, and broadcast through which these updates change predictions across prefixes and reshape the rollout distribution. This framework helps us understand both limited student responsiveness despite substantial teacher-student disagreement, and degeneration when updates that improve local teacher agreement also increase the probability of unsuccessful continuations, exposing the student to unreliable supervision in subsequent rollouts. Finally, we show that offline distillation can prepare a student for OPD without significantly improving immediate held-out accuracy. While OPD alone may degrade generation, it can yield substantial accuracy gains after sufficient offline preparation. Together, our results connect teacher supervision, student geometry, and the evolving rollout distribution in determining the learning dynamics of OPD and its effects on downstream performance.
Nov 5, 2026

CoRD: Coupled Rollouts Fill the Blind Spot of On-Policy Distillation

On-policy distillation trains a student on its own rollouts, so states the teacher would reach but the student does not are seldom supervised. We introduce CoRD (coupled-rollout distillation), which keeps the on-policy loss and moves only the distribution of training states toward the teacher, drawing rollouts from a geometric interpolation of student and teacher through the maximal coupling of speculative decoding. The coupling keeps the student's token wherever the two distributions agree and otherwise substitutes a token from the part of the target the student underweights, so interventions are as rare as any exact sampler allows and recede as the student learns. Across six base students from two model families (Qwen3 and Qwen3.5, 0.6B to 4B), CoRD raises average pass@1 over on-policy distillation by 0.9 to 2.0 points and pass@8 by 1.2 to 3.4 points on five public math benchmarks at equal steps and data. On frozen teacher-reached states, a 4B student trained with CoRD reaches in 16 steps a gap closure that on-policy distillation does not reach in 80, and substitution rules intervening at least as often in other directions fall short of CoRD.
Nov 5, 2026

O2^2PD: Overlap-Gated On-Policy Distillation on Reliable Prefix

On-policy distillation (OPD) supplies dense token-level supervision by aligning a student to the teacher's next-token distribution on its own generated rollouts. Existing works mostly implement top-kk OPD, applying reverse Kullback-Leibler (\RKL\RKL) divergence between the teacher's and the student's renormalized distributions over teacher top-kk support. However, the teacher scores student-generated rollouts that are inherently off-policy for itself, leaving the reliability of its next-token guidance unclear. Restricting the divergence to the teacher top-kk support also discards the remaining mass outside that support. These gaps leave two questions implicit: 1) which tokens to align, and 2) how to align those tokens. To answer these questions, we propose Overlap-gated On-Policy Distillation (O2^{2}PD). As for 1), we quantitatively analyze the reliability of teacher supervision and find that alignment on the student's trajectory prefix is more reliable than alignment on the full context. Therefore, our O2^{2}PD is applied only to the 20% prefix tokens of each student rollout. As for 2), we theoretically characterize two problems of top-kk OPD: the Teacher-support Uncovering and the Dominant-Mode Mismatch. Based on these analyses, we propose two support strategies: the Reset Support strategy attaches a reset atom to every token, pooling mass outside the teacher top-kk, and the Support Convergence strategy converges the support to the already aligned intersection. Our O2^{2}PD employs the designed overlap-gated αα-divergence to interpolate between mode-covering and mode-seeking for per-token alignment, where the overlap is used as a gate to switch the support and adaptively modulate αα.
Nov 5, 2026

Reinforcing What the Student Has Learned: Teacher-Guided On-Policy Distillation

On-policy distillation (OPD) improves language-model reasoning through dense teacher feedback on student-generated responses. We first observe an amplitude disparity in this feedback: on math reasoning responses, style-related positions exhibit 3.5 times the mean absolute teacher--student log-probability difference of mathematical positions. Scaling OPD updates by these discrepancies thus directs much of the update strength toward matching the teacher's expression. We instead take the student's own pretrained base as a reference: the student--base difference records the token preferences acquired during post-training and provides supervision that does not depend on the student's distance from the teacher. Building on this reference, we propose Student-Delta On-Policy Distillation (SD-OPD), which uses centered student--base differences as update coefficients, keeps each coefficient only when its sign agrees with centered teacher--student feedback, and recomputes the student signal every training round. Across 5 teacher--student configurations with Qwen3 students of 1.7B to 8B parameters, SD-OPD improves six-task mathematics average pass@1 (avg@16) over OPD by 7.76 to 10.37 percentage points. Code is available at https://github.com/aim-uofa/sd-opd.
Nov 4, 2026

D2D^2-OPD: Separating Direction and Distance in Extrapolative On-Policy Distillation

On-policy distillation (OPD) trains a student model to match a teacher's next-token distributions on student-generated prefixes. Extrapolative OPD further extends the teacher's log-probability shift relative to a reference policy to construct stronger learning targets. However, existing methods couple extrapolation direction and distance, allowing the magnitude of the teacher's shift to dictate how far the target moves. We propose D2D^2-OPD, decoupling these two factors: a per-prefix hard KL constraint bounds the target's distance from the teacher, while an adaptive combination of the teacher's post-training shift and the initial student--teacher gap determines the extrapolation direction. We optimize the student toward this target using forward KL, providing learning signals even for low-probability tokens. We evaluate three settings: where standard OPD performs well, where it struggles with base-model initialization, and where no teacher reference is available. On mathematical reasoning benchmarks, D2D^2-OPD attains the highest score in all three settings, outperforming the strongest baseline by up to 9.45 points. Ablations validate the contributions of each component.
Nov 4, 2026

EvoOPD: Evolving Privileged Information Strategies for Black-Box On-Policy Distillation

Proprietary language models exhibit strong reasoning and coding capabilities, but their output distributions are inaccessible, making fine-grained distillation methods that rely on teacher logits difficult to apply directly to black-box models. We observe that rich textual feedback from a black-box teacher can be treated as privileged information (PI) and converted, through a PI-conditioned privileged teacher, into dense supervision signals on the student's own trajectories. However, what PI should be acquired from the teacher to most effectively support student learning remains unclear. PI strategies that differ in where they focus, what information they request, and how much detail they provide can have substantially different effects on the current student, and their relative effectiveness can further change as training progresses. Based on this observation, we propose Evolving Privileged Information Strategies for Black-Box On-Policy Distillation (EvoOPD), which treats PI acquisition itself as an optimization problem and maintains a PI Strategy Bank that evolves with the student state. EvoOPD uses the measured repair utility of PI on the current student to adapt the focus, information type, and granularity of teacher feedback, allowing the acquired information to continually match the student's learning needs. We systematically evaluate EvoOPD on mathematical reasoning and code generation with Qwen3-1.7B, 4B, and 8B students. EvoOPD achieves the best overall performance in all six model-task settings, improving over the strongest baseline in each setting by an average of 2.14 percentage points in mathematics and 1.30 points in code. Further analysis shows that the relative effectiveness of PI acquisition strategies changes with student state, making a fixed strategy difficult to sustain as the best choice throughout distillation.
Nov 4, 2026

Ruby-OPD: Enhancing Pronunciation Controllability and Accuracy for Flow-Matching Text-to-Speech via On-Policy Distillation

Achieving both high pronunciation accuracy and fine-grained controllability remains a central challenge in Japanese text-to-speech (TTS), as Japanese is an unsegmented language mixing logographic Kanji and phonographic Kana with extensive polyphony. Traditional cascaded systems convert text into phoneme sequences via a front-end, ensuring pronunciation control but discarding lexical semantics and prosodic cues in surface orthography. Conversely, recent zero-shot TTS models directly consume raw orthographic text to improve naturalness, yet suffer from polyphone errors and lack explicit pronunciation control. To bridge this gap, we propose a two-stage framework for flow-matching TTS. First, we investigate Ruby text representations with orthography perturbation to jointly preserve surface forms and target pronunciations for reliable control. Second, we introduce On-Policy Distillation (OPD) on large-scale text-only corpora to enhance pronunciation accuracy on raw orthographic text without paired speech. On the Joyo-Kanji-Yomi benchmark, our model achieves 97.13% Accuracy, 2.02% Target PER, 0.70% Sentence PER, and 4.21% Sentence CER under raw text while retaining strong controllability.
Nov 3, 2026

On the Generalization Limitation of On-Policy Distillation for Post-RL Consolidation

Multi-teacher on-policy distillation (MOPD) extends on-policy distillation (OPD) to consolidate capabilities from multiple teachers. Although prior work reports near-perfect in-distribution (IND) distillation performance, we find that this can mask incomplete capability consolidation beyond the supervised distribution. A preliminary study on synthetic reasoning shows that OPD does not fully transfer teachers' generalization capabilities. We then extend this investigation to real-world tasks, evaluating extrapolative and compositional generalization across mathematical reasoning, coding, agent, and long-context capabilities using Qwen3 models. Across extrapolative settings, OPD and MOPD reach 96.04-102.73% of their teachers' IND scores. Yet their out-of-distribution (OOD) scores remain 2.45-16.25% below the base model's. Across three compositional settings, MOPD improves IND performance over the base model, but its performance on composed tasks as OOD remains 18.19-24.34% below that of the best teachers. To understand these gaps, we examine teacher capability and data coverage as two potential bottlenecks. Jointly alleviating these bottlenecks eliminates OOD forgetting in our experiments and suggests the potential for generalization beyond the teacher. (M)OPD also reduces OOD forgetting on average relative to its reinforcement learning (RL) baselines. Analysis of the associated update geometry reveals smaller and more spectrally concentrated updates on average than those of the RL baselines, along with greater shared update structure under MOPD. Motivated by these findings, we propose geometry-based patching and base-preserving replay to improve OOD performance while largely preserving IND performance. Our findings identify a blind spot in capability consolidation via (M)OPD, highlighting the need for algorithms that explicitly support OOD retention and generalization.
Nov 3, 2026

Stabilizing On-Policy Targeted Guidance in Agentic Reinforcement Learning

Naive on-policy RL is often bottlenecked by the capabilities of the base model --- insufficient coverage can result in an inability to learn from certain rewards. At the same time, rewards in modern post-training systems often come from rich textual sources, e.g. judge models with a rubric of criteria. Can we utilize such structures to provide a rich directional signal that enables improving on otherwise difficult to optimize rewards? In this work, we examine a framework we refer to as targeted guidance that learns from per-turn textual feedback via an on-policy self-distillation objective. Our focus is not only on directly improving hard-to-optimize behaviors, but on how to do so throughout the training process without degrading baseline RL performance. Over a range of settings across an agentic multi-turn software engineering environment, we find success by combining targeted self-distillation with a per-criteria RL reward, and by limiting feedback to turns where the policy specifically fails. We further examine the design choices that lead to stable entropy and reduce oscillatory behavior throughout training.
Nov 2, 2026

P2^2-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors

Multi-teacher on-policy distillation consolidates expert capabilities by supervising student-generated trajectories, whereas task arithmetic offers a training-free alternative through linear parameter combinations. We view the linear combination of student and expert parameters as a single differentiable model and propose P2^2-MOPD, which jointly optimizes the mixing coefficients and the student weights under the same on-policy distillation objective. Unlike task arithmetic, which stays within the span of the frozen student and expert parameters for any choice of merging coefficients, P2^2-MOPD keeps the student trainable, so the experts act as parameter priors without confining it to this span. Across four domains at 1.7B and 4B scales, P2^2-MOPD achieves faster and more balanced recovery than MOPD from the same initialization, surpassing its final average recovery in half the training steps. At the end of training, our 1.7B model reaches 94.7% average recovery, versus 80.9% for MOPD.
Nov 2, 2026

Meet Students Where They Are: Student-Prefix Teacher Adaptation for On-Policy Distillation

On-policy distillation (OPD) mitigates the student's exposure bias by optimizing on student-generated trajectories. However, student-generated prefixes may contain reasoning errors or follow patterns unfamiliar to the teacher, undermining the reliability of its token-level supervision. This degradation becomes more pronounced as prefixes grow longer. To address this teacher-side distribution mismatch, we propose Student-Prefix Teacher Adaptation (SPTA), which adapts the teacher to the states on which it is queried during distillation. Specifically, before OPD starts, we collect student rollout prefixes with random lengths and continually train the teacher to continue from the prefixes using reinforcement learning with verifiable rewards (RLVR). By treating each prefix as context and rewarding correct final answers, the teacher learns to recover from student-induced states rather than imitate the student's reasoning. Then, the adapted teacher is used for standard OPD, with no changes to the distillation objective. Experiments on mathematical reasoning benchmarks show that our SPTA improves average student performance over standard OPD and OPD with task-level RLVR-adapted teachers. These results highlight the importance of adapting teachers not only to solve problems independently, but also to provide reliable guidance on the states their students actually visit.
Nov 2, 2026

SpecOPD: Branch-Parallel On-Policy Distillation for Speculative Decoding

Speculative decoding (SD) accelerates large language model inference by reducing the sequential cost of token-by-token generation. Advanced SD methods reduce proposal cost with parallel drafters that predict entire blocks in one forward pass, but overall speedup still depends on acceptance length. On-policy distillation (OPD) can push well-trained drafters beyond acceptance-length plateaus, yet conventional pipelines collect supervision through sequential draft-and-verify rollouts. Our key insight is that on-policy supervision need not inherit these dependencies: a draft block can be supervised using only its reference prefix and draft-sampled history. We propose SpecOPD, a branch-parallel OPD framework that parallelizes draft sampling and target supervision across independent branches drawn from the evolving drafter policy at reference anchors. We pack these branches into proposal trees to share target-prefix computation, using tree masks for full attention, branch-consistent indexer groups for sparse attention, and indexed state replay for linear attention to preserve each branch's history. We train the drafter with an E2E-TV surrogate that directly targets acceptance length. Our study spans 8B full-attention (GQA), 27B full-linear (GQA--GDN), and 125B sparse-linear (QSA--GDN) targets, three parallel draft architectures (DFlash, DSpark, and DFlash V2), and four task domains (dialogue, coding, mathematics, and agentic). SpecOPD achieves up to 41.45% higher mean acceptance length, 16.09×16.09\times supervised-token throughput over sequential-rollout OPD under native configurations, and up to 3.56×3.56\times generation speedup over autoregressive decoding.
Nov 2, 2026

OmniOPD: Stable On-Policy Distillation for Joint Audio-Video Generation

Reinforcement learning (RL) elevates joint audio-video generation, but optimizing heterogeneous, modality-specific audio and video rewards often introduces cross-modal interference. On-policy distillation (OPD) alleviates this issue by transferring the capabilities of specialized audio and video RL teachers into a single student model. However, a significant imbalance in cross-modal token counts leaves audio with sparse supervision, causing unstable distillation. We observe that naively increasing the audio loss weight fails to solve this problem since audio learning relies heavily on visual context, indicating richer visual cues are crucial to offset sparse audio signals. To address this, we propose OmniOPD, a modality-wise on-policy distillation framework designed for balanced and stable audio-video generation. First, we construct privileged visual information using asymmetric noise scheduling, allowing the audio teacher to generate more informative targets from cleaner video context. Second, we introduce grouped multi-noise sampling to expand audio supervision coverage and stabilize training. Third, we introduce timestep-wise importance weighting based on teacher-student transition divergence, preserving reliable supervision while down-weighting unreliable learning signals, to stabilize fluctuations in audio learning. Evaluated on JavisBench and VBench demonstrate OmniOPD preserves audio quality while significantly enhancing video generation, achieving effective joint distillation.
Nov 1, 2026

Order-Only On-Policy Distillation

On-policy distillation (OPD) trains a student on its own generations by matching the teacher's token probabilities. These probabilities entangle two kinds of information: which tokens the teacher prefers, and how confident it is. The latter reflects the teacher's own capacity and calibration, which a smaller student may be unable to reproduce. Because the teacher temperature changes these values without changing the ordering, OPD is also sensitive to the temperature choice. We ask whether the teacher's ordering of candidate tokens is enough, and propose Order-Only On-Policy Distillation (O3^{3}PD). At each student-generated prefix, O3^{3}PD trains the student to reproduce the order of the teacher's top-KK tokens, without fitting their probability values. The objective is therefore invariant to the teacher temperature and needs no temperature tuning. Across six teacher--student pairs, O3^{3}PD outperforms OPD on the average over HMMT25, AIME24, and AIME25 in every pair, by up to 2.9 points, with gains of up to 5.6 points on AIME25. Other ranking losses yield similar gains, so the benefit comes from ordinal supervision itself. Analysis shows that O3^{3}PD departs further from the teacher's probabilities than OPD, yet reproduces its ranking more faithfully and yields a stronger student. The ranking, not the probability values, thus carries the knowledge that on-policy distillation needs to transfer.
Nov 1, 2026

Bernoulli On-Policy Distillation: Don't Ignore High-Confidence Student Tokens

On-policy distillation (OPD) trains a student language model on its own trajectories using token-level supervision from a stronger teacher. We show that standard sampled reverse-KL OPD systematically suppresses updates to high-confidence tokens, such that their weak gradient contribution does not necessarily imply weak or uninformative teacher supervision. We introduce Bernoulli On-Policy Distillation (Bernoulli OPD), which directly matches the teacher and student probabilities for that token. This yields a bounded probability space correction that preserves teacher-student disagreement while avoiding the additional confidence-dependent suppression induced by vanilla OPD. Fixed-rollout analyses show that Bernoulli OPD substantially restores optimization signal to high-confidence tokens while remaining sensitive to teacher-student disagreement. We further analyze the resulting model behavior through confidence distributions, answer calibration, policy movement, and training dynamics, showing that Bernoulli OPD's gains cannot be explained simply by longer reasoning, larger optimization steps, or more uniform imitation of the teacher. Code is available at https://github.com/duterscmy/Bernoulli-OPD.
Nov 1, 2026

Better Answers, Weaker Evidence Control: Evidence Bypass in On-Policy Distillation

On-policy distillation (OPD) can improve answer accuracy while weakening responsiveness to supporting evidence. We investigate this separation in scientific-document question answering through matched document interventions that change the supported answer while preserving the question and unrelated context. We identify evidence bypass: a student retains an originally correct answer after the document changes to support a different one. On QASPER with a Qwen3-4B student, standard OPD raises original-document accuracy by 3.4 percentage points, lowers counterfactual accuracy by 5.3 points, and increases bypass by 15.7 points; the same directional pattern appears on PeerQA. Controlled perturbations localize the shift to answer-critical evidence, and bypass grows with the student's pre-distillation closed-book preference for the original answer. A factorized study of teacher targets, privileged information, divergence, and supervision density finds that explicitly supplying the teacher with annotated answer-critical spans yields the largest joint improvement in evidence response and claim support. These findings motivate Evidence-Paired OPD (EP-OPD), which combines on-policy supervision on matched documents supporting different answers to the same question with an answer-level margin that favors each document's supported answer over its paired alternative. Across both datasets and two student sizes, EP-OPD recovers 54-66% of standard OPD's counterfactual-accuracy loss while preserving 80-85% of its original-document accuracy gain. Our results establish evidence responsiveness as a distinct training and evaluation target alongside answer accuracy. Our code is publicly available at: https://github.com/RuoranXu/EP-OPD.