Language Model Distillation

Follow this topic Ask about this topic

Momentum

69 papers in the last four weeks, up 200% on the four weeks before. 0.7% of all new papers.

Jul 13Week of Sep 28

Latest papers 350

Aug 5, 2026cs.AI

When Privileged Guidance Misaligns: State-Matched Routing and Contextualized Self-Distillation for Multi-Turn Agents

Privileged on-policy distillation provides dense supervision for multi-turn agents by allowing a synchronized teacher to re-score the student's response at every turn with access to training-only references, such as successful trajectories. In interactive environments, however, the student's preceding actions continually change the execution state. As the student takes different actions or completes subgoals in a different order, its rollout may reach states not covered by the reference, making the reference an unreliable source of guidance for the state actually reached. Applying privileged distillation indiscriminately therefore creates state--reference mismatch. This mismatch motivates a central objective: providing privileged reference guidance that remains compatible with the student's current execution state. We introduce State-Matched Routing and Contextualized Self-Distillation (SMRC-SD), which explicitly determines when and how a privileged trajectory should guide an on-policy student. At each turn, SMRC-SD verifies whether the student's current execution state matches a supported state along the reference trajectory. Distillation is applied only at matched states, filtering out turns for which the reference lacks locally compatible guidance. For each matched state, SMRC-SD further constructs state-conditioned teacher context from the successful trajectory, grounding supervision in the state actually reached. Across ALFWorld and WebShop, SMRC-SD consistently outperforms unconditional successful full-path distillation. With Qwen3-1.7B, it improves task success from

0.746

to

0.865

on ALFWorld and from

0.574

to

0.693

on WebShop. Controlled routing and context ablations support both selecting locally supported turns and constructing state-compatible teacher context as contributors to these gains. Code is available at https://github.com/liujunzhuo/SMRC-SD.

Junzhuo Liu, Weiwei Li, Jun Ling +1

University of Electronic Science and Technology of China

Language Model Distillation On-Policy Distillation Knowledge Distillation RL for LLM Agents

Aug 5, 2026cs.LG

SPOT: Sparse Probing and Outcome Calibration for On-Policy Distillation

On-policy distillation (OPD) provides dense teacher supervision on student-generated trajectories, but standard reverse-KL training can assign insufficient probability to other plausible continuations. Teacher entropy alone does not reveal whether uncertainty is concentrated among a few plausible next tokens or dispersed over a long probability tail, nor whether the student already represents those candidates well. Moreover, local teacher probabilities may not predict downstream success. We introduce Sparse Probing and Outcome-calibrated Targets OPD (SPOT), which addresses two coupled decisions, where to probe and what to distill, through an acquisition--exploration--exploitation procedure. During acquisition, a position-level score combines normalized teacher entropy, the probability mass captured by a small top-

k

candidate set, and student--teacher mismatch to allocate a limited probing budget. During exploration, SPOT evaluates teacher-proposed candidates through verifier-scored student continuations. During exploitation, these outcomes produce a closed-form, KL-regularized target that favors candidates with better downstream outcomes while remaining anchored to the teacher distribution. Extensive experiments across multiple student models and reasoning benchmarks demonstrate the effectiveness of SPOT in improving reasoning performance while balancing solution quality and coverage.

Zikun Qu, Min Zhang, Mingze Kong +4

The Chinese University of Hong Kong, Shenzhen · East China Normal University · Beihang University +1

Language Model Distillation On-Policy Distillation Knowledge Distillation

Aug 5, 2026cs.LG

Not Every Divergence Should Be Suppressed: Counterfactual Recoverability in On-Policy Distillation

On-policy distillation (OPD) supervises student-visited trajectories, yet divergence-based rules cannot determine whether an erroneous prefix remains correctable. We formulate this decision as counterfactual recoverability and replay each error state through budget-matched teacher-continuation and rollback branches. Based on their relative success, states are categorized as recoverable, irreversible-but-avoidable, or ambiguous, and these labels guide whether training retains, rolls back, or conventionally supervises the corresponding trajectory. On AIME branch diagnostics, the mean continuation-minus-rollback effect is 0.185 for recoverable states and -1.000 for irreversible-but-avoidable states, demonstrating opposite intervention preferences. A branch-derived recoverability proxy achieves an AUC of 1.000, substantially outperforming divergence alone at 0.392. Across frozen evaluations, recoverability-aware control achieves the strongest recorded performance, reaching 0.578 success on held-out AIME2025 compared with 0.517 for the best baseline. It also improves AIME2024-2025 average@32 from 0.2656 to 0.3125 and GPQA-Diamond average@32 from 0.2702 to 0.3070. Component ablations further show that retaining teacher-correctable prefixes provides the largest individual contribution. These findings establish recoverability as an outcome-grounded decision variable for selective supervision in OPD.

De Jiang, Zhengyang Zhang, Kehong Yuan +1

Language Model Distillation On-Policy Distillation Counterfactual Learning

Aug 4, 2026cs.CL

Efficient Knowledge Distillation for LLMs: Offline Top-K Logits and a Fused Chunked KL Loss

Small language models are often the only option for deployment under tight latency, cost, and on-premises constraints, but they are rarely trained from scratch: a compressed model is usually recovered through knowledge distillation (KD). This recovery step largely decides the final quality, yet it is expensive. We present a practitioner's study of how to make distillation training efficient, organised around two systems contributions. First, we show that offline KD (caching the teacher's top-

K

logits once and training the student against the cache) matches online distillation at near-identical training loss while removing the teacher from memory, running about 29% faster per iteration, and reaching up to 41% higher throughput on a single H200 GPU. Second, we introduce a \emph{fused, chunked KL loss} that never materialises the full vocabulary-sized logit tensor, making peak memory linear in the sequence length. This removes the memory spike that otherwise caps context length and lets us train at four times the context (32{,}768 tokens) on a single GPU. A separate output-head-only toy benchmark isolates the loss kernel and confirms its memory and iteration-rate scaling from 4K to 256K tokens. Together these make large-scale healing and hundreds of ablations affordable. We also report supporting ablations on loss design and sequence packing. We release our chunked-loss implementation: https://github.com/CompactifAI/Full-Chunked-KL-Loss.

Bakbergen Ryskulov, Iker García-Ferrero, David Montero +5

Multiverse Computing

Efficient Language Model Training Language Model Distillation Knowledge Distillation

Aug 4, 2026cs.AI

When Teachers Mislead: Spurious-Signal-Aware On-Policy Distillation

On-Policy distillation (OPD) transfers teacher capabilities by supervising student-sampled trajectories with dense token-level teacher signals. Recent selective OPD methods improve this process by prioritizing signals that are confident, informative, or learnable. However, the assumptions overlook a fundamental failure mode of language models: their token-level judgments can be driven by input-agnostic language priors, formatting conventions, or stereotyped reasoning templates rather than task-specific evidence. We refer to such optimization-relevant but weakly input-grounded supervision as spurious signals in OPD, which may produce large gradients while contributing little task-improving direction. To mitigate this issue, we propose SA-OPD, a Spurious-Signal-Aware On-Policy Distillation framework that identifies and filters misleading token-level supervision based on input-groundedness and optimization impact. SA-OPD introduces a lightweight input-groundedness proxy estimating whether a token-level distillation signal truly depends on the input. It then filters only tokens that simultaneously exhibit low input-groundedness and extreme distillation divergence, thereby removing high-impact spurious updates and achieving fine-grained OPD optimization. Extensive experiments on both large language model (LLM) and vision-language model (VLM) settings demonstrate that SA-OPD consistently outperforms Vanilla OPD and competitive selective methods. These results establish input-groundedness as a key dimension for OPD supervision selection and offer a simple, effective strategy for mitigating spurious updates.

Yinuo Jiang, Yongjie Ye, Zhou Tao +4

Zhejiang University · ByteDance · Shanghai Artificial Intelligence Laboratory

VLM Distillation Language Model Distillation On-Policy Distillation Self-Supervised Knowledge Distillation Knowledge Distillation

Aug 4, 2026cs.CL

TQLite: Multi-LLM Jury Guided Distillation for Real-time MQM Translation Quality Evaluation

Large language models (LLMs) have demonstrated impressive performance in MQM-based translation quality (TQ) evaluation, and recent advances in large reasoning models (LRMs) promise even greater improvements. However, both LLMs and LRMs are computationally expensive to deploy at scale, while small language models (SLMs)---though much more efficient---struggle with the complex reasoning required for evaluation tasks. In this work, we present an extensive empirical study benchmarking SLMs, LLMs, and LRMs across a wide range of TQ evaluation setups, providing a comprehensive view of the current landscape and establishing best practices. To address the scalability challenge, we introduce TQLite, a novel distillation framework that enables SLMs to approach the MQM evaluation performance of the best LRM-based evaluators. Our approach leverages a multi-LRM jury to generate high-quality synthetic training data via practical data curation techniques and aggregation of evaluation responses across a diverse panel of models. Our results demonstrate that SLMs trained via TQLite achieve strong MQM evaluation performance that far exceeds off-the-shelf evaluation capabilities of standard SLMs, offering a scalable and cost-effective alternative to LLM- and LRM-based evaluators.

Bhavin Jawade, Cameron R. Wolfe

Netflix

LLM Evaluation LLM-as-a-Judge Small Language Models Language Model Distillation MT Evaluation

Aug 3, 2026cs.CL

Stuck on "A": Diagnosing and Repairing Interface Injury in Attention-to-KDA Linearization of a 0.6B Language Model

We convert 21 of 28 full-attention layers of Qwen3-0.6B-Base into KDA (Kimi Delta Attention) linear-attention layers on a single consumer-grade GPU budget, and ask a simple question: what exactly does the conversion break? After surgery, hidden-state alignment and end-to-end KL distillation drive the student close to its teacher in perplexity, yet multiple-choice accuracy stays near random chance (25-29% vs. the teacher's 50.6% on C-Eval). Using a four-permutation diagnostic that rotates answer options while holding content fixed, we show the model sticks to option labels (predicting "A" 81% of the time; 106/161 questions keep the same label under all four rotations) rather than following answer content -- an interface injury that standard distillation metrics cannot see. A 1,000-step format-targeted completion-only KL stage repairs the interface (+12.48 points on C-Eval, label-stickiness roughly halved), after which persona SFT and one round of on-policy DPO preserve benchmark scores within noise. We release code, weights, recipes, and the full audit trail, and distill the engineering lessons -- including an FP32-master failure mode in which bf16 optimizer updates are silently swallowed -- that made convergence possible at this budget.

Ronglong Bao

DT-Project

Delta-Rule Linear Attention Language Model Distillation LLM Post-Training

Aug 3, 2026cs.AI

PCSD: Persistent Consistency for Self-Distillation in Agentic Reinforcement Learning

Large language model agents have shown strong potential in complex interactive tasks, yet their reinforcement learning (RL) is often hindered by sparse rewards, as a long multi-turn trajectory may receive only a single outcome-level signal. On-policy self-distillation (OPSD) provides dense token-level supervision from a privileged teacher, but the teacher may not be reliable at every position. Existing methods commonly rely on isolated token-level discrepancies, which can be sensitive to noise, or assign a shared step-level weight that may overlook positional variation. We propose Persistent Consistency Self-Distillation (PCSD), which derives token-level distillation weights from the local persistence of teacher-favoring signals. PCSD combines adaptive windows with exponentially decayed aggregation to capture persistent relative teacher support, applies trend-aware modulation to attenuate locally declining support, and produces continuous weights through sigmoid gating. The resulting objective is jointly optimized with GRPO, combining dense teacher guidance with sparse environmental feedback. Without inference-time skills, PCSD achieves the best ALFWorld Overall results among all baselines on both backbones, exceeding GRPO by 15.6 and 13.3 points and SDAR by 6.2 and 5.5 points, while remaining competitive on WebShop and gaining 15.8 points over GRPO on unseen ALFWorld split.

Chunji Lv, Yangguang Wei, Junlin Liu +6

Beijing Institute of Technology · Meituan · Institute of Automation, Chinese Academy of Sciences +1

Language Model Distillation On-Policy Distillation Sparse-Reward RL RL for LLM Agents

Aug 3, 2026cs.AI

DAPD: Dual-Anchored Policy Distillation

On-policy (self) distillation (OPSD) is increasingly adopted for language-model post-training. It strengthens the teacher with privileged information but can induce a privilege illusion: the student learns privilege-dependent behavior it cannot reproduce from its inference-time context, yet behaves as if the training-time privileged information remained available, ultimately degrading performance. In this paper, we identify information asymmetry between the privileged teacher and the student at inference as the root cause of this failure in OPSD. To resolve this asymmetry, we propose Dual-Anchored Policy Distillation (DAPD), a unified framework with two levels of anchoring. Dual-Path Anchoring (DPA) introduces a self-conditioned bridge and aligns reference and rollout behavior along two matched-information paths, preventing privilege-dependent behavior from being transferred to the inference-time student. Dual-Source Anchoring (DSA) applies these paths in both reference-to-rollout and rollout-to-reference directions, reducing reliance on privileged reference guidance while preserving correctness supervision. Extensive experiments show that DAPD significantly alleviates privilege illusion, outperforming OPSD on Qwen3-4B by +2.00 points on average across tasks. Notably, its gains persist across scales, reaching +2.69 at 4B and +2.78 at 32B.

Jianyu Wu, Yizhou Wang, Encheng Su +2

Shanghai Jiao Tong University · Shanghai Artificial Intelligence Laboratory · The Chinese University of Hong Kong +1

On-Policy Self-Distillation Language Model Distillation RL Post-Training Policy Distillation

Aug 1, 2026cs.CL

Native Multilingual Chain-of-Thought Reasoning in Low-Resource Southeast Asian Languages

Large Language Models have achieved substantial progress in reasoning capabilities. Yet in low-resource native settings, many suffer from cross-lingual collapse, reverting to English during intermediate steps that require complex logical reasoning. This presents a cold-start bottleneck for policy optimization, whereas standard fine-tuning risks catastrophic forgetting due to cross-lingual representation drift. To address these challenges, we introduce the Onramp-Sequence Cross-Distillation (OSCD), a post-training algorithm that projects high-resource reasoning trajectories into low-resource vocabulary subspaces during generative training rollouts via an integrated translator agentic loop, ensuring the stable and efficient translation of dynamically generated reference samples for fine-tuning. This is coupled with joint-embedding semantic alignment of both reference and target-language reasoning traces, thereby bridging the pairwise cross-lingual representational gaps. Comprehensive evaluations using the AIME25 and HMMT25 benchmarks demonstrate that OSCD yields up to 3.2 times overall improvements in native Southeast Asian languages for mathematical reasoning, of which the joint-embedding semantic alignment component contributes up to 6.4% improvements in linguistic debiasing over translation-only baselines.

Sean Gip Lim, William Chandra Tjhi, Hai Leong Chieu

1Nanyang Technological University · 2AI Singapore · 3DSO National Laboratories

Numerical Reasoning in Language Models Multilingual Language Models Cross-Lingual Reasoning in Language Models Cross-Lingual Representation Alignment Language Model Distillation Low-Resource Language Processing+1

Jul 31, 2026cs.LG

Adaptive FastOPD: Progress-Aware Rollout Horizon Expansion for Efficient On-Policy Distillation

On-policy distillation (OPD) provides dense teacher supervision along student-generated trajectories, but its online rollout process incurs substantial computational cost, particularly when a few long responses delay batch completion. Existing acceleration methods typically control rollout length using fixed budgets or absolute teacher--student agreement thresholds, which may not reflect learning progress across different models and training stages. We propose Adaptive FastOPD, a progress-aware strategy that expands the rollout horizon only when learning near the current boundary region has plateaued and the current horizon is sufficiently utilized. The former is determined from four teacher--student signals measured relative to their values upon entering each horizon, making expansion responsive to stage-specific progress rather than a predefined step interval or an absolute threshold on the raw agreement signals, while the latter prevents a small number of long responses from triggering increases in rollout cost. Across two teacher--student pairs, Adaptive FastOPD achieves the highest average performance while reducing training time by 49.1--71.2% relative to OPD 15K, and remains robust across a range of hyperparameter settings.

Qian Tan, Huaifei Liang, Xuanyu Zhu +2

University of Science and Technology of China · 3Shanghai Jiao Tong University · 2Shanghai Artificial Intelligence Laboratory

Language Model Distillation On-Policy Distillation

Jul 30, 2026cs.LG

Distilling Knowledge from Large Language Models into Lightweight Reinforcement Learning Agents for Autonomous Cyber Operations

Autonomous Cyber Operations (ACO) are increasingly important for defending enterprise networks as cyber threats continue to evolve in sophistication. ACO applications commonly employ Reinforcement Learning (RL) agents to learn defensive behaviors through interaction with environments. However, RL agents typically require extensive exploration during training, often resulting in unstable behavior and poor initial decision-making before converging toward effective defense strategies. In this work, we investigate the use of a Large Language Model (LLM) to improve autonomous defensive decision-making within an ACO environment. Through prompt engineering rather than fine-tuning, we demonstrate that an 8-billion parameter LLM pretrained on cybersecurity data can outperform a baseline RL agent in a modified CybORG CAGE Challenge 2 environment. We then propose an online policy distillation framework that transfers the LLM's defensive policy into a lightweight RL agent containing only 64,910 parameters, reducing model size by several orders of magnitude while maintaining effective defensive capabilities. This provides a pathway toward operationalizing frontier cybersecurity models within lightweight, deployable agents. To evaluate transferability, we construct CybORG scenarios ranging from 4 to 12 hosts and assess the approach across varying network configurations. We also evaluate teacher-guided RL stabilization strategies and observe that none consistently surpass the optimized teacher policy, suggesting policy-alignment limitations between reward-driven RL optimization and teacher-guided defense strategies. Our results demonstrate the potential of cybersecurity-focused LLMs as sources of expertise for autonomous cyber defense, while policy distillation provides a practical path toward operationalizing frontier cybersecurity models within efficient, scalable agents.

Konur Tholl, François Rivest, Mariam El Mezouar +2

Department of Electrical and Computer Engineering, Royal Military College of Canada, Kingston, Canada · Department of Mathematics and Computer Science, Royal Military College of Canada, Kingston, Canada · Defence Research and Development Canada, Ottawa, Canada +1

Autonomous Cyber Defense LLM-Guided RL LLMs for Cybersecurity Language Model Distillation

Jul 30, 2026cs.LG

$β$ -OPSD: Deriving with Policy Optimization, Training with Self-Distillation

On-policy self-distillation (OPSD) is a promising approach to improve reasoning language models, but it remains brittle in practice: making it work reliably often requires substantial engineering effort. We identify a structural source of this difficulty: vanilla OPSD is precisely the

β=1

member of a broader policy-optimization family, where

β

weights the KL penalty anchoring the student to a reference policy. This equivalence turns

β

from an implicit value fixed at one into a controllable regularization parameter, yielding a more general formulation that trades off proximity to a reference policy against privileged teacher guidance. We introduce

β

-OPSD and derive its optimal policy as a geometric interpolation between the reference policy and the privileged teacher. Directly optimizing this objective with reinforcement learning, however, would be costly and high-variance. Rather than optimize the RL objective directly, we turn its closed-form solution into a distillation target. Each value of

β

selects a target along the reference-to-teacher path, which we implement efficiently by mixing their token-level logits. In this way, inexpensive distillation approximates the solution of expensive policy optimization. Return-to-go credit assignment further aligns token updates with the sequence-level objective while retaining the simplicity of OPSD. Experiments on mathematical reasoning benchmarks show that

β

-OPSD consistently outperforms vanilla OPSD, improving optimization stability and downstream reasoning performance. Our results provide a principled route from self-distillation to policy optimization and back without sacrificing the efficiency that makes OPSD practical.

Jiawei Xu, Minghui Liu, Juzheng Zhang +2

University of Maryland, College Park

KL-Regularized RL On-Policy Self-Distillation RL for Language Model Reasoning Language Model Distillation Policy Learning

Jul 30, 2026cs.CL

Lightning OPD 2.0: Mitigating Style Bias in Cross-Teacher On-Policy Distillation for Large Reasoning Models

On-policy distillation (OPD) provides dense token-level supervision from a teacher, but its effectiveness can depend on teacher consistency, meaning that the model providing OPD supervision should also have generated the demonstrations used to train the supervised fine-tuning (SFT) reference. However, this condition is frequently violated in practice when SFT data have mixed or unknown provenance or when different models are preferred for SFT data generation and subsequent distillation. In such cross-teacher settings, even a stronger OPD teacher can yield little improvement over the SFT reference. We find that raw teacher--reference disagreement contains potentially useful context-specific teacher evidence as well as a recurring component associated with differences in wording, formatting, and reasoning cadence. We introduce Lightning OPD 2.0 with cross-fitted style residualization, which uses rollout-level cross-fitting to estimate this recurring component as an operational proxy for style-token bias and subtracts it before constructing the token-level OPD update. Across mathematical reasoning and code generation benchmarks, Lightning OPD 2.0 consistently outperforms Lightning OPD in cross-teacher settings. Starting from Klear-Reasoner-8B-SFT, Lightning OPD 2.0 reaches 82.4% on AIME 2024 and 63.0% on LiveCodeBench v5. Together, these results establish Lightning OPD 2.0 as a practical approach to cross-teacher OPD, relaxing teacher consistency as a prerequisite and allowing the SFT data generator and distillation teacher to be selected independently. Code will be released soon.

Yecheng Wu, Song Han, Han Cai

NVIDIA

Language Model Distillation On-Policy Distillation Knowledge Distillation LLM Reasoning

Jul 30, 2026cs.LG

Flux-OPD: On-Policy Distillation with Evolving Contexts

Large language model training in open-ended domains lacks verifiable rewards, making task preferences difficult to formalize as effective supervision. Contexts can convey such preferences, yet provide little additional supervision once distilled into the student, motivating contexts that evolve with student performance. However, directly using evolving contexts as in-training supervision results in an unstable distillation target and conflicting distributions, requiring mechanisms to stabilize target and downweight conflicts. In this paper, we analyze the effect of contexts through a decomposition of the reverse KL objective, revealing two findings: the student is distilled toward the geometric mean of context-conditioned teachers, and the objective contains a conflict term that measures conflicts among these teachers. Based on this decomposition, we propose Flux-OPD, an OPD paradigm that uses evolving contexts as in-training supervision to capture task preferences in open-ended domains. Flux-OPD treats the differences between context-conditioned and context-free teachers as contextual difference signals, injects them as contextual corrections into the context-free teacher anchor, and weights their correction strength using the conflict term as an indicator. Experiments on open-ended tasks show that Flux-OPD outperforms existing OPD paradigms, highlighting the potential to combine teacher supervision with evolving contexts.

Yuran Wang, Zekun Wang, Bohan Zeng +10

Peking University · Kling Team · Tsinghua University +3

Language Model Distillation On-Policy Distillation Context Distillation Knowledge Distillation

Jul 28, 2026cs.LG

Weak-to-Strong On-Policy Distillation

On-policy distillation (OPD), which aligns a student with the teacher's token-level distribution on the student's own rollouts, is an effective paradigm for transferring capabilities across LLMs. Prevailing approaches assume a teacher at least as capable as the student: they either distill a larger model into a smaller one, which fails at the frontier where no larger teacher exists, or consolidate multiple domain experts trained from a shared base, which requires costly training at the student's scale. We introduce Weak-to-Strong On-Policy Distillation (W2S-OPD), a simple yet effective OPD framework that improves the strong student by distilling from multiple weak models. W2S-OPD constructs a proxy teacher in logit space from a contrast pair of a positive and a negative model, both smaller than the student and cheap to obtain. Their logit difference isolates the capability direction, which is added to the student's own base model, yielding a proxy teacher that couples this direction while staying distributionally adjacent to the student. The student then distills it by minimizing the per-token reverse KL on its own rollouts. We instantiate the contrast pair as i) a post-RL expert against its pre-RL initialization, isolating the skill RL instills, ii) a larger against a smaller base model, isolating the capability from scale, and iii) a small base model with correct versus wrong hints, isolating the instance-level direction toward the solution. Across four math and three code benchmarks, W2S-OPD outperforms OPD, enables the student to surpass the domain teacher, and keeps improving the student even when every supervision source is weaker. Analysis shows different contrasts yield distinct signals: the post-RL and hint contrasts emphasize reasoning frameworks, while the scale contrast emphasizes the solving procedure. Our code will be available at https://github.com/Yu-Fangxu/W2S-OPD.

Fangxu Yu, Weijia Xu, Michael Xu +2

University of Maryland, College Park · Microsoft Research · MBZUAI

Language Model Distillation On-Policy Distillation Knowledge Distillation

Jul 28, 2026cs.CL

Pass the Baton: Trajectory-Relayed On-Policy Distillation

On-policy distillation (OPD) grounds token-level supervision in the student's own trajectory, yet suffers from prefix failure: once the student commits to a wrong reasoning direction, all subsequent generation builds on this deviation, producing misdirected continuations that elicit unreliable supervision and waste compute. We identify a teacher-student continuation asymmetry on failed prefixes, where the teacher tends to redirect while the student continues along the original direction, and convert it into a label-free handoff trigger in Relay On-Policy Distillation (Relay-OPD). During training, Relay-OPD constructs relay trajectories by letting the teacher briefly take over at detected trigger points to produce a teacher leg, after which the student resumes and is optimized on the resulting trajectory. A limited relay budget concentrates intervention on critical early positions while limiting departure from the student policy. With a Qwen3-4B-Instruct-2507 teacher and Qwen3-0.6B/1.7B-Non-Thinking students on eight mathematical reasoning benchmarks, Relay-OPD achieves the best or second-best results on every benchmark, outperforming standard OPD by +5.73% and the strongest baseline FastOPD by +1.49% on average for 1.7B, with consistent gains at 0.6B. Training trajectory length is reduced by over 50%.

Haolei Xu, Xiaowen Xu, Haiwen Hong +5

1Zhejiang University · 2Yuvion Team, Alibaba Group

Language Model Distillation On-Policy Distillation Policy Learning

Jul 26, 2026cs.LG

Outcome-Confounded Local Supervision in On-Policy Distillation

On-policy distillation (OPD) trains a student on its own trajectories while a teacher supplies dense token-level likelihoods at student-visited prefixes. These likelihoods are often read locally: agreement appears safe to imitate, whereas disagreement appears to identify an error. We show that both readings are confounded by the outcome of the completed trajectory. We introduce an outcome-resolved diagnostic that crosses pointwise teacher-student divergence with final-answer correctness, separating safe imitation, productive divergence, harmful divergence, and agreement-on-failure. In an eight-seed mathematical-reasoning study with a Qwen3-8B student and Qwen3-32B teacher, agreement-on-failure constitutes 67.84% of pooled response-token mass; with a Qwen2.5-7B/32B pair it remains 67.68%. The result persists across threshold, sequence-level, format, and truncation audits. Even on prompts that the Qwen3 teacher solves in all four independent attempts, student accuracy rises to 86.91% but agreement-on-failure remains 14.76%. We then run three matched training probes that use the available signals to imitate, mask, or contrast whole trajectories; none consistently reduces agreement-on-failure. The result points to a localization limitation: local divergence paired with a trajectory-level outcome does not identify where a failed trajectory became unrecoverable. Addressing this limitation requires additional positional information, such as process labels, teacher continuations from student prefixes, or token-level alignment across rollouts. Our contribution is therefore diagnostic rather than a new training method.

Guoqing Ma

The Chinese University of Hong Kong, Shenzhen

Credit Assignment in RL Language Model Distillation On-Policy Distillation

Jul 24, 2026cs.LG

Cross-Tokenizer On-Policy Distillation via Byte-Prefix Marginalization

Open-weight language models from different families exhibit complementary capabilities, motivating their consolidation into a compact student through on-policy distillation (OPD). However, full-vocabulary OPD typically assumes a shared tokenizer, while existing cross-tokenizer methods may discard teacher probability mass or assign it to student tokens with unrelated content. We introduce Byte-Prefix Marginalization (BPM), which re-expresses the teacher's next-token distribution over the student vocabulary in a shared byte space. Specifically, BPM assigns each teacher token's probability to the longest student token whose byte representation is a prefix of the teacher token's bytes, aggregates mass mapped to the same student token, and places otherwise unmatched mass in an explicit residual category. This produces a vocabulary-complete, byte-aligned, and mass-preserving target for dense OPD. The target exactly recovers the teacher-induced byte-prefix marginal when the relevant prefix does not span multiple teacher tokens (a condition satisfied at more than 99% of training positions) and uses a mass-preserving, chain-factorized lower bound otherwise. Across Qwen3-32B, GLM-Z1-9B-0414, and MiniMax-M2.7 as teachers, BPM consistently outperforms current cross-tokenizer methods on six mathematics and programming benchmarks, improving six-benchmark avg@8 by 3.7-6.6 points over the strongest baselines.

Hao Wang, Kun Yuan, Wenlin Zhong +4

University of Chinese Academy of Sciences · KwaiKAT Team · Zhejiang University +1

Cross-Tokenizer Knowledge Distillation LLM Compression Language Model Distillation On-Policy Distillation Knowledge Distillation

Jul 23, 2026cs.LG

X $^3$ -OPD: Distilling Reasoning into Large Audio-Language Models via On-Policy Alignment

While large audio-language models have achieved remarkable progress in auditory perception, they still lag behind text-based large language models in deep logical reasoning, primarily due to the scarcity of high-quality audio reasoning data. To bridge this gap, we propose X

^3

-OPD, a cross-modal on-policy distillation framework that transfers reasoning capabilities from a powerful text teacher to an audio-language student. During training, the student generates reasoning trajectories conditioned on its own acoustic perception, while the teacher provides token-level guidance using matched textual inputs and verified answers. We further construct a three-tier symmetric corpus covering textual reasoning rendered into speech, audio-event reasoning grounded in complex acoustic scenes, and spoken-dialogue reasoning involving paralinguistic cues. This design extends cross-modal distillation beyond textually recoverable content to reasoning grounded in non-linguistic events, prosody, and conversational context. Experiments on MMSU, MMAU, BIG Bench Audio, and MMAR demonstrate that X

^3

-OPD substantially improves audio-grounded reasoning and chain-of-thought quality while largely preserving the model's existing capabilities under domain shift.

Dongjie Fu, Di Cao, Xize Cheng +6

1Tencent Hunyuan · 2Zhejiang University

Cross-Modal Knowledge Distillation Audio Reasoning Audio-Language Models Language Model Distillation On-Policy Distillation

Jul 23, 2026cs.CL

Sample-Efficient Learning from Agent Experience

Real-world agent learning is often constrained by costly environment interactions, such as running time-consuming experiments or obtaining human feedback. In-context learning offers a highly sample-efficient way for agents to learn from their own interaction histories, but its gains disappear once that experience is removed from the context. Separately, context distillation provides a mechanism for internalizing contextual information into model weights. However, applying it to agents' interaction histories without sacrificing environment sample efficiency remains underexplored. We term this problem Experience Distillation and develop an implementation that requires no further environment interaction beyond the collected experience. Experiments on 749 curated software-engineering tasks and six text-adventure games show that it retains at least 64.8% of the gains from in-context learning across both domains, whereas direct supervised fine-tuning on the collected experience recovers only 3.8%. Compared with classical reinforcement-learning baselines, in-context learning from trial-and-error experience followed by Experience Distillation matches their performance with at least

9.6\times

fewer environment samples.

Chenhui Gou, Haoqin Tu, Yunhao Fang +2

Monash University · ByteDance Seed

In-Context RL Language Model Distillation Context Distillation Sample-Efficient RL RL for LLM Agents

Jul 21, 2026cs.LG

REGEN: Replay-recycling for Expert-to-Generalist distillation with Offline Reinforcement Learning

Large-scale online reinforcement learning (RL) is the predominant means of eliciting advanced abilities including long-term reasoning and agentic tool use in large language models (LLMs). However, continuing to scale it across vast task domains of interest remains challenging in both computational infrastructure and cost, especially when considering RL as merely a one-off learning stage. Recently, a widely used technique for distilling knowledge across various domains and training stages, multi-teacher on-policy distillation (MOPD), helps to decouple the RL stage, saving costs, while maintaining generality across vast domains. Nonetheless, similar to online RL, MOPD requires coupled inference and backward passes, which continues to limit its scalability and computational efficiency. To address these challenges, we propose REGEN: Replay-recycling for Expert-to-Generalist Distillation with Offline RL. Instead of distilling from multiple teacher models, REGEN trains a generalist by simply recycling the replay memory -- the free by-product of the teachers' specialized RL training -- and employing offline RL algorithms. REGEN completely decouples the rollout sampling from the backward training process and thus greatly reduces the training cost. Across mathematical reasoning, code generation, and instruction following, REGEN matches the accuracy of MOPD at substantially lower cost. It potentially turns online RL into a data synthesis process instead of a one-off learning stage, and can be extended to large-scale post-training without requiring heavy computational load. Code is available at https://github.com/yunjie-sysu/REGEN.

Yunjie Chen, Xiaoxin Chen, Fang Wang

vivo AI Lab, Shenzhen, China · Department of Computer Science, Sun Yat-Sen University, Guangzhou, China

RL for Language Models RL for Language Model Reasoning Offline RL Language Model Distillation RL Post-Training

Jul 21, 2026cs.CV

Contrastive On-Policy Distillation

On-policy distillation (OPD) trains a student model on trajectories sampled from its own policy, providing dense token-level supervision by minimizing the divergence between the teacher's and student's output distributions at each token. Although existing OPD approaches effectively distill strong reasoning capabilities into student models, they inherently inherit uncurated chain-of-thought traces, exacerbating overthinking and reasoning redundancy. To address this limitation, we introduce COPD, a contrastive OPD framework. Specifically, for each token generated by the student, a frozen teacher evaluates the current student state under two contrasting prompts that induce low and high reasoning effort, respectively. The resulting difference in log-probabilities serves as a token-level advantage signal to guide the policy update. Rather than merely mimicking a single teacher distribution, COPD guides the student model toward acquiring concise and efficient reasoning strategies. We evaluate COPD across 9 multimodal benchmarks covering both reasoning and understanding tasks. Empirical results show that COPD substantially reduces reasoning length without hurting task performance, consistently improving efficiency across different tasks and model scales. In addition, this contrastive paradigm extends naturally to on-policy self-distillation (OPSD), establishing self-contrastive supervisory signals that enable a single model to compress its own reasoning without an external teacher.

Jiacheng Ruan, Jun Tang, Wenzhen Yuan +5

Shanghai Jiao Tong University · Alibaba Group

Language Model Distillation On-Policy Distillation Knowledge Distillation

Jul 21, 2026cs.LG

H $^2$ SD: Hybrid Hindsight Self-Distillation

Reinforcement learning with verifiable rewards (RLVR) provides reliable outcome supervision for language model reasoning, but a scalar trajectory reward offers limited token-level guidance. Existing self-distillation methods add a privileged teacher but typically assign it a fixed role: direct distribution matching may destabilize successful behavior, while magnitude-only modulation offers little corrective guidance after failure. We observe that successful and failed trajectories require different forms of hindsight supervision. A successful response already contains a valid student-generated reasoning path and can therefore serve as privileged context rather than being replaced by an external rationale. A failed response, however, requires corrective reference information. We introduce Hybrid Hindsight Self-Distillation (

\mathrm{H}^{2}\mathrm{SD}

), which jointly adapts teacher context and update strategy to trajectory correctness. For successful trajectories, we construct the teacher context from the verified response and a rephrasing instruction, and use the teacher only to re-evaluate the original response tokens. The resulting probabilities refine token credit assignment without changing the direction determined by the reward. For failed trajectories, a verified reference hint provides corrective guidance through reverse-KL distillation. Experiments on challenging reasoning benchmarks show that H

^2

SD achieves the strongest overall performance among representative RLVR and self-distillation baselines, with stable optimization and a favorable accuracy-efficiency trade-off.

Qiye Cai, Yichuan Ma, Peiji Li +6

1Shanghai Artificial Intelligence Laboratory · 2Harbin Institute of Technology · 3Fudan University +1

RL for Language Model Reasoning Language Model Distillation Reinforcement Learning with Verifiable Rewards Self-Distillation

Jul 21, 2026cs.CL

Rationale-Guided Knowledge Distillation for Cross-Lingual Stance Detection

Stance detection aims to identify whether a text expresses a favorable or opposing attitude toward a given target, and serves as an important task for various downstream applications. Although existing studies have achieved strong performance in monolingual settings, especially in English, many low-resource languages such as Catalan still lack sufficient annotated data for training effective models. Cross-lingual stance detection alleviates this problem by transferring stance knowledge from resource-rich languages to low-resource languages. However, most existing methods mainly rely on semantic alignment between texts and targets, while ignoring the reasoning process required for reliable stance inference. Although Large Language Models provide strong reasoning ability, their high computational cost and inference latency limit practical deployment. To address these limitations, we propose a rationale-guided knowledge distillation framework for cross-lingual stance detection. Specifically, we use Chain-of-Thought prompting to guide Large Language Models in generating informative rationales, and distill the resulting reasoning knowledge into a compact student model. We further design a dual-path distillation mechanism to align rationale-enhanced and rationale-free representations, together with their prediction distributions. In addition, two contrastive learning strategies are introduced to improve stance discrimination. Experiments on multilingual benchmarks demonstrate that our method consistently outperforms competitive baselines.

Qiuli Zhou, Jingyuan Yao, Shengeng Tang +3

School of Foreign Studies, Hefei University of Technology, China · School of Computer Science and Information Engineering, Hefei University of Technology, China · School of Computer Science and Information Engineering, Hefei University of Technology +1

Stance Detection Language Model Distillation Low-Resource Language Processing Cross-Lingual Transfer

Jul 18, 2026cs.CL

Trace-Based On-Policy Distillation for Masked Diffusion Language Models

Diffusion large language models (dLLMs) are a promising alternative to autoregressive generation. However, reasoning-oriented post-training for dLLMs remains challenging. Supervised fine-tuning (SFT) for dLLMs requires dense but often off-policy masked states, while reinforcement learning (RL) relies on sparse rewards or value modeling. This paper proposes \textbf{trace-based on-policy distillation (TOPD)}, a teacher-supervised framework that transfers reasoning ability to a target dLLM without reward estimation. The key idea is to supervise a dLLM on its own denoising trajectory, focusing on the trace-aligned token decisions that form the final response. Specifically, TOPD samples on-policy diffusion trajectories from the target dLLM, obtains teacher token distributions from a teacher model on the corresponding partially denoised states, and updates the target dLLM with a token-level Reverse Kullback-Leibler (Reverse-KL) objective. This design preserves dense teacher supervision while aligning training with the model's own denoising states. On mathematical reasoning benchmarks, TOPD enables SDAR-4B-Chat to match the MATH500 accuracy of its RL-trained counterpart TraDo-4B-Instruct, with gains of +5.7 under static evaluation and +4.5 under dynamic evaluation. Compared with the RL-trained counterpart, TOPD achieves this with 4

\times

fewer rollout rounds, corresponding to an estimated 96.0

\times

to-accuracy model-compute speedup.

Haolin Ren, Ziyang Huang, Chenhao Yuan +2

Institute of Automation, Chinese Academy of Sciences, Beijing, China · University of Chinese Academy of Sciences, Beijing, China

Masked Diffusion Models Diffusion Model Distillation Language Model Distillation Diffusion Language Models LLM Mathematical Reasoning Diffusion Language Model Distillation

Jul 16, 2026cs.LG

On-Policy Delta Distillation

On-policy distillation is an alternative post-training method in reinforcement learning that alleviates the constraints imposed by reward models by providing token-level supervision from a teacher model. Although on-policy distillation has been studied and applied across various settings, its fundamental design remains underexplored. In this paper, we introduce a new distillation reward, termed the delta signal, instead of directly imitating the teacher's output distribution. The delta signal is defined as the difference between the teacher model and its base model prior to instruction tuning for reasoning capability. It therefore captures the changes induced by reasoning tuning and provides a more direct signal for transferring reasoning capabilities. Using extensive empirical evidence, we show that the delta signal substantially improves on-policy distillation and refer to the new distillation method as On-Policy Delta Distillation (OPD

^2

). Experiments across mathematics, science, and code-reasoning benchmarks demonstrate that OPD

^2

consistently outperforms conventional on-policy distillation, enabling reasoning LLMs to achieve strong performance with only a short post-training period. Code will be available at https://github.com/naver-ai/opd2

Byeongho Heo, Jaehui Hwang, Sangdoo Yun +1

NAVER AI Lab

RL for Language Model Reasoning Language Model Distillation On-Policy Distillation

Jul 16, 2026cs.CL

Gold-Guided Programmatic Distillation for Financial Reasoning over Hybrid Tables and Text

Financial question answering over hybrid tabular and textual data may require multi-source reasoning and precise numerical computation. While large language models (LLMs) can generate intermediate reasoning steps, natural-language rationales remain prone to arithmetic errors, making them an unreliable supervision source for distillation. Building on programmatic distillation, we develop an approach that transfers reliable numerical reasoning from a large teacher model to a compact student using execution-verified Python programs instead of free-form textual rationales. It leverages gold derivations to guide teacher-side program synthesis and retains only programs that execute correctly and produce the gold answer, ensuring high-quality supervision. We further introduce an iterative recovery stage that revisits teacher-failed examples, enabling the student to recover and incorporate newly verified programs into training. Experiments on TAT-QA show that our framework is highly effective for hybrid financial reasoning. Our best 7B student achieves 87.00 EM / 87.18 F1 on the test set, substantially outperforming the 72B teacher (78.46 EM) as well as traditional and strong LLM-based baselines, including TAGOP and TAT-LLM. These results demonstrate that execution-verified programmatic distillation provides an effective and extensible framework for training smaller models to perform reliable numerical reasoning.

Yun Dong, Erica Zhao, Elana Chen

College of Computing · Georgia Institute of Technology · Department of Management Science & Engineering +2

Table QA Financial QA LLM-Based Program Synthesis Mathematical Reasoning Language Model Distillation Finance

Jul 16, 2026cs.CL

Answer-Conditioned Chains of Thought Degrade Verifiable-Reasoning Distillation in Large Language Models

A standard recipe for distilling the reasoning ability of large language models (LLMs) is to sample chains of thought from the model, keep those that reach the correct final answer, and fine-tune on the survivors. When sampling fails, a common fix shows the generator the gold answer and asks it to write a chain that reaches that answer. We show that this second step degrades the training data in a way that correctness filtering cannot catch. We run a controlled experiment that fixes the generator, the problem set, and the correctness filter, and varies only whether the chain is generated under answer-conditioning, the gold answer shown with a request to reach it. Training a strong instruction-tuned reasoning model on its own answer-conditioned chains sharply lowers its verifiable-reasoning accuracy. The loss grows with difficulty, reaching as much as about 27 points on the hardest competition problems. The mechanism is legible in the chains themselves, which rationalize backward from the shown answer instead of deriving it, with the early final-answer statement as the measurable symptom. The harm is a property of the data rather than the generator, read off unlabeled generations before any fine-tuning, ordering the penalty across eight thinking models from four families, and transferring across teacher families. A prompt ablation localizes it to the rationalize-toward instruction rather than the answer's bare visibility. The practical takeaway is to generate answer-blind, because no correctness filter can see this damage in the data.

Jungseob Lee, Seungyoon Lee, Suhyune Son +4

1Korea University · 2Zoom Communications · 3Yonsei University

CoT Distillation Language Model Distillation LLM Reasoning Reasoning Verification

Jul 15, 2026cs.LG

Consensus as Privileged Context for Label-Free Self-Distillation

Sampling multiple solutions and returning the majority answer is among the most reliable ways to improve the reasoning accuracy of large language models without labels, and a growing family of methods converts this consensus signal into training supervision. However, existing approaches use consensus only in restricted forms: as a filter that selects solutions for fine-tuning, as a preference between answers, or as a scalar reward for reinforcement learning, discarding most of the information that the agreeing solutions contain. We present CANON (Consensus-ANchored self-distillatiON), a label-free training method that turns consensus into dense, token-level supervision. For each unlabeled prompt, CANON samples multiple solutions, extracts the majority answer, and conditions a frozen snapshot of the model on a solution that reaches it; this consensus-anchored teacher then supervises the model on its own rollouts at every token. Experiments on mathematical and scientific reasoning benchmarks show that CANON improves pass@1 by up to 12 points, outperforming label-free reinforcement learning by 6 points at a seventh of its compute and approaching a teacher conditioned on gold solutions; trained on pooled unlabeled data, it transfers to held-out benchmarks, matching training methods that use gold labels. Analysis suggests that the improvements are not pure distribution sharpening: after training, the model solves problems it previously never solved in 32 attempts, and its majority vote itself becomes more accurate.

John Gkountouras, Josip Jukić, Ivan Titov

1ILLC, University of Amsterdam · 2ILCC, University of Edinburgh

Self-Training for Language Models Language Model Distillation Self-Distillation Scientific Reasoning in Language Models Language Model Self-Improvement LLM Reasoning