Policy Distillation

Follow this topic Ask about this topic

Momentum

7 papers in the last four weeks, against 2 the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 27

Oct 7, 2026cs.LG

Composing What Each Teacher Learned: Multi-Teacher On-Policy Distillation through Teacher-Relative Shifts

Multi-teacher on-policy distillation (MOPD) is used in two settings. In common-domain composition, several teachers score each student rollout from one prompt domain and their signals form a single target; in routed-domain distillation, prompts from different domains are assigned to the corresponding specialist. Both settings usually transfer each teacher's endpoint policy, which mixes what post-training changed with preferences inherited from the teacher's base. We introduce

Δ

-MOPD, which transfers each teacher's teacher-minus-base logit shift re-anchored at the student's frozen initialization, and compare it with endpoint supervision in both settings while holding teacher selection fixed. We first expose the mechanism that impedes endpoint transfer: inherited base pull can exceed the post-training shift. Removing it reduces the teacher-term norm ratio and target--student KL. Across our experiments, the results suggest that shift targets are particularly useful when teacher signals are combined at a state. With three composed teachers,

Δ

-MOPD exceeds endpoint composition by

4.11

Math and

1.95

five-benchmark points; with two, it matches endpoint accuracy. Under phased routing, it achieves higher mean performance in both phase orders and reduces the observed order gap from

10.50

to

6.42

points. Under interleaved routing, where each update involves one teacher, the two targets perform comparably. The phased results provide supporting evidence that the benefit may extend to signals accumulated across training phases. Target construction is thus an independent design axis in MOPD, complementary to teacher selection.

Hejian Sang, Zhengze Zhou, Shayan Mohajer Hamidi +3

Iowa State University · LinkedIn · Harvard University

Multi-Teacher Knowledge Distillation On-Policy Distillation Policy Distillation

Oct 6, 2026cs.RO

iGPC: Generative Motion Priors for Object-Aware Humanoid Interaction

Humanoid robots operating in unstructured environments must combine robust whole-body control with the ability to perceive and physically interact with surrounding objects. While large-scale human motion data provides powerful priors for natural and versatile humanoid control, effectively transferring such priors to perception-driven object interaction remains challenging. To address this bottleneck, we propose a framework that extends the recently proposed Generative Pretrained Controller (GPC) from general human motion to full-body humanoid-environment interaction. First, we adapt GPC into interaction experts conditioned on scene affordance cues and privileged state information. These experts leverage the pretrained human motion prior while learning task-specific contact behaviors, including reaching toward objects, grasping environmental supports for stabilization, and pushing movable objects. Second, we introduce a perception-driven student that retains the pretrained GPC policy and distills interaction skills from the experts using onboard sensory observations. To bridge the gap between privileged expert observations and sensory inputs, we propose two complementary training objectives that enable effective adaptation of the pretrained motion prior during distillation. Notably, our experiments across multiple whole-body interaction tasks demonstrate that large-scale generative human motion priors provide an effective foundation for learning deployable policies for humanoid interactions in contact-rich real-world environments.

Anujith Muraleedharan, Abdul Ahad Butt, Nolan Fey +5

Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), Abu Dhabi, United Arab Emirates. · Massachusetts Institute of Technology (MIT), Cambridge, MA, USA. · Minerva Humanoids, Inc.

Contact-Rich Robotic Manipulation Humanoid Robot Control Whole-Body Robotic Manipulation Generative Modeling Policy Distillation Human-Object Interaction

Sep 30, 2026cs.RO

Linear Recurrent Memory Suffices to Distil a World-Model Policy for Robot Air Hockey

Does memory-dependent control need nonlinear recurrent dynamics? We study simulated air-hockey defence under temporary loss of puck tracking. A DreamerV3 teacher outperforms a memoryless policy under tracking loss, while resetting the teacher's recurrent state sharply reduces performance, which demonstrates that the task requires memory. We distil this teacher into compact recurrent policies with a 64 dimensional state, with a combination of a diagonal linear recurrence and an optional rank-

k

nonlinear innovation while retaining nonlinear observation encoders and action heads. Across five matched seeds, the purely linear recurrent model (

k=0

) matches both the GRU baseline and the teacher throughout the tested range of tracking loss. Increasing nonlinear innovation rank providing no measured benefits. This result is obtained on a fresh test split, which will be only opened after all models and analyses are frozen. The linear model requires fewer recurrent parameters and less computation than GRU, but performs comparably. These results suggest that, for this memory dependent control task, nonlinear representation learning around a simple linear memory mechanism can be sufficient, and that nonlinear recurrent dynamics are not necessarily required. These conclusions are limited to the simulated task, teacher, state dimension, and blackout horizon considered here, and to policies whose observation encoder and action head remain nonlinear.

F. Olivia Fan, Oliver Obst

UNSW Sydney, NSW 2052, Australia

Recurrent Neural Networks Robotic RL Linear RNNs Policy Learning Policy Distillation

Sep 29, 2026cs.LG

Interpolated Policy Distillation: A Controllable Continuum Between Off-Policy and On-Policy Distillation

Off-policy and on-policy distillation have traditionally been formulated as separate paradigms, each favoring a different property of distillation trajectories. Teacher-generated (off-policy) traces are typically high-quality but lie far from the student's distribution, whereas student-generated (on-policy) rollouts are more learnable but often contain erroneous reasoning. We view these paradigms as the endpoints of a policy continuum and posit that a more effective rollout policy may lie in between. We introduce \textbf{Interpolated Policy Distillation (IPD)}, which defines the next-token distribution at every decoding step as an explicit linear interpolation between the student and teacher distributions. The interpolation operates at the distribution level, token by token, and its coefficient provides direct control over the balance between trajectory quality and student learnability. Naively sampling from this policy would require sequentially querying the teacher at every token and is thus expensive. To make IPD practical, we accelerate it with a new speculative-decoding rule while exactly preserving the interpolated next-token distribution.At the trajectory level, the resulting rollouts naturally interleave student- and teacher-generated segments. Unlike recent heuristic segment-interleaving methods, however, this interleaving is induced by an exactly realized token-level interpolated policy rather than by hand-designed switching rules. Across text-only and multimodal reasoning benchmarks, IPD consistently outperforms both endpoint policies (SFT and OPD), their conventional two-stage combination (SFT-then-OPD), and recent heuristic segment-interleaving methods, demonstrating that token-level policy interpolation better balances trajectory quality and student learnability.

Youxu Shi, Yifan Sun, Dacheng Yin +5

University of Science and Technology of China · WeChat Vision, Tencent Inc.

Language Model Distillation On-Policy Distillation Policy Distillation Knowledge Distillation

Sep 29, 2026cs.RO

Distilling Privileged Control Barrier Functions into RGB-Only Safety Filters for Dynamic Visual Navigation

RGB-only end-to-end visual navigation policies remain vulnerable to collisions in real-world dynamic environments, motivating a dedicated safety layer. Existing visual Control Barrier Function (CBF) approaches seek to provide safety from RGB observations, but often rely on real-time rendering or explicit scene reconstruction and are primarily designed for static scenes, limiting their practicality for onboard deployment. We propose a teacher-student visual distillation framework that transfers the safety behavior of a privileged CBF teacher to an RGB-only student filter for dynamic environments. The student maps a short RGB history, robot velocity, and a nominal control action directly to a safe action, while the teacher uses ground-truth robot and obstacle states in a real-to-sim dynamic Gaussian Splatting environment. To reduce the teacher-student information gap, the teacher constructs safety constraints only from obstacles observable within the student's RGB history. It also accounts for obstacle-velocity uncertainty to improve robustness to motion variations, while action augmentation exposes the student to diverse safe and unsafe nominal actions to better capture the safety boundary. At deployment, the student requires only RGB observations and robot velocity, without explicit 3D reconstruction or online rendering. Experiments show that the proposed method outperforms visual CBF baselines and improves the safety of RGB-based navigation policies under dynamic obstacle motion. Project page: https://syeon-yoo.github.io/distill-cbf-site/.

Seungyeon Yoo, Gawon Lee, Seungwoo Jung +2

Department of Aerospace Engineering, Seoul National University, Seoul, Republic of Korea

Control Barrier Functions Robotics Safe Robot Navigation Robot Safety Dynamic Obstacle Avoidance Vision-Based Robot Control+1

Sep 28, 2026cs.LG

Inductive Feedback for Mixed-Policy Distillation

Verbal feedback can identify errors and prescribe corrections, providing rich supervision for language-model post-training even when reliable programmatic verifiers are unavailable. Such feedback, often generated by a capable model, can be used to condition the teacher in on-policy distillation, which trains the student to match the teacher's predictions on student-generated rollouts. However, this approach can transfer teacher preferences that the feedback did not motivate, while leaving much of the feedback's guidance unused. We find that both problems come from the standard on-policy distillation objective, specifically the divergence it minimizes and the distribution it uses as its target. Our proposed method addresses both limitations. First, to isolate the information conveyed by the feedback from the teacher's inherent preferences, we treat verbal feedback as evidence for or against the hypothesis that a particular token comes next at a given prefix. We then adopt a probabilistic confirmation framework which uniquely determines an ordering over the vocabulary based on the teacher's predictions before and after it receives feedback. Using a confirmation score consistent with this ordering, we construct a target distribution within a trust region of the student. Second, to learn from guidance that student rollouts can leave unused, we derive a simple shared-rollout estimator of a symmetric divergence between the student and target distributions over rollouts, reusing student and feedback-conditioned teacher rollouts in both directions through importance weighting. Empirical evaluations show that our method outperforms the common on-policy distillation recipe and a recent contrastive variant on knowledge-based and agentic benchmarks.

Amir Moeini, Huaijiang Zhu, Daniel Havir +1

Alquist Robotics · University of Virginia

Language Model Distillation On-Policy Distillation Policy Distillation

Sep 28, 2026cs.CL

Beyond Token Alignment: Event Completion for Cross-Tokenizer On-Policy Distillation

On-policy distillation (OPD) transfers knowledge between language models through teacher supervision on student-generated trajectories. With different tokenizers, a single teacher token may require multiple student tokens to generate, creating intermediate states where the event is entered but not yet completed. Existing cross-tokenizer methods align tokens or text spans to construct comparable prediction targets. We study a complementary problem after partial generation: once the student produces a prefix of a teacher token, multiple next tokens may complete the same remaining bytes, but the teacher only specifies the required completion rather than how probability should be divided among these valid continuations. We introduce Event-Set Completion Distillation (ESCD), which complements cross-tokenizer probability alignment with completion-set supervision. ESCD aggregates prefix-related teacher events and supervises the total probability of byte-compatible one-step student completions, avoiding tokenizer-dependent probability splits among individual tokens. The method reuses student trajectories and predictions, requiring neither additional rollouts nor changes to the student vocabulary. Experiments demonstrate consistent gains in mathematics, code, and scientific reasoning across model families and tokenizers, extending to large-scale MoE distillation from a 1T teacher to a 35B student. Local analyses show that retaining completion sets better matches the reference supervision, while one-step completion covers over 99% of observed compatible teacher mass after partial event entry in the studied tokenizer pairs. These findings support event entry and event completion as complementary supervision targets for cross-tokenizer knowledge transfer. Code will be released on GitHub.

Jiacheng Liu, Jingwei Song, Qituan Zhang +2

Shanghai Jiao Tong University · Fudan University

Cross-Tokenizer Knowledge Distillation On-Policy Distillation Policy Distillation

Sep 20, 2026cs.RO

STRIDER: Stepping-Enabled Multi-Gait Hierarchical 3D Loco-Manipulation Framework for Humanoid Robots

Humanoid loco-manipulation faces two prominent limitations: controllers using continuous velocity commands cannot precisely regulate individual footholds, while specialized foothold-tracking modules are difficult to integrate with whole-body manipulation. Furthermore, standard action-based imitation distillation primarily transfers expert actions, without explicitly encouraging a shared representation of heterogeneous skills. This paper introduces STRIDER, a hierarchical multi-gait framework to bridge these gaps. The framework integrates terrain-aware 3D stepping logic, Adversarial Motion Priors (AMP)-based natural walking, and Cartesian upper-body control: its stepping expert selects feasible footholds in the stance-foot frame and generates clearance-aware swing trajectories. To fuse distinct walking and stepping experts into one executable student policy, we propose Latent Distillation Proximal Policy Optimization (LD-PPO), a distillation algorithm augmented with teacher-conditioned latent alignment. By jointly optimizing on-policy reinforcement learning, DAgger-based action reconstruction, and latent alignment, LD-PPO transfers expert actions while encouraging a shared skill representation across heterogeneous modes. Simulation and real-robot evaluations on the TianGong Omni humanoid show that LD-PPO outperforms vanilla distillation-PPO in foothold-tracking and posture-tracking accuracy. Deployed on hardware, STRIDER realizes multi-gait loco-manipulation with accurate foothold and end-effector tracking.

Yuanzhuo Li, Wen Zhao, Zhe Yong +6

X-Humanoid, Humanoid Innovation Department

Humanoid Loco-Manipulation Humanoid Robot Locomotion Robot Imitation Learning Policy Distillation

Sep 8, 2026cs.CL

Experience Funnel: A State-Policy Alternating Loop for Self-Evolving Agents

Autonomous agents powered by large language models (LLMs) continuously accumulate experience through interaction, creating an opportunity to improve future behavior through self-evolution. A fundamental challenge is how to transform abundant, task-specific interaction experience into reusable model competence without sacrificing the ability to adapt rapidly to newly observed evidence. Explicit textual states, such as skills and agent harnesses, provide fast, human-readable and editable adaptation, but incur persistent dependence on external context; parametric policies provide compact and reusable competence, but are substantially slower to update. We present \textit{Experience Funnel}, a self-evolving framework that couples fast state adaptation with slow policy consolidation in an alternating loop. Interaction trajectories are first distilled into an explicit textual state, where newly acquired experience can be rapidly incorporated and validated. The framework then selectively identifies state-enabled behavior that remains useful across state revisions and consolidates it into the policy through transition-aware distillation. The updated state--policy pair subsequently generates new rollouts, providing fresh evidence for the next round of state adaptation and policy consolidation. Experiments across diverse agent benchmarks show that \textit{Experience Funnel} consistently improves agent capability over state-only evolution and policy-internalization approaches, while progressively converting useful explicit experience into autonomous policy competence.

Wenbo Gao, Zhaomou Song, Zhiyuan Ji +7

1The Hong Kong Polytechnic University · 2Huawei · 3Renmin University of China

Continual Learning for LLM Agents Agent Skill Learning Policy Distillation Self-Evolving Agents

Sep 1, 2026cs.AI

Selective Agent Guidance via Entropy: Learning Autonomous Policies from Imperfect VLM Teachers

Vision-Language Models (VLMs) provide useful priors for interactive decision-making, but using them directly as policies is expensive and brittle: they must be queried at every step, do not improve from environment interaction, and can repeat systematic errors. We study how to learn a cheap autonomous policy from an online, expensive, and imperfect but informative VLM teacher. We propose SAGE (Selective Agent Guidance via Entropy), a framework that queries a VLM only when the learner is uncertain, executes the suggested action during training, and distills guidance into a lightweight Reinforcement Learning (RL) policy. Because VLM advice is not always reliable, SAGE can weight teacher-action distillation using environment-derived advantages rather than treating all suggestions as equally useful. Across sparse-reward visual reasoning and navigation tasks, SAGE learns policies that act without VLM guidance at evaluation time and improves over unguided RL in several environments, including settings where the learned policy exceeds its VLM teacher. The results show that selective guidance is most beneficial when the VLM can help the agent discover high-reward trajectories, and less useful when unguided exploration already succeeds or teacher actions do not lead to informative experience. SAGE also reduces VLM usage by prompting the teacher only on a fraction of training steps and requiring no VLM calls at deployment. Overall, our results suggest that VLMs don't need to be used as fixed policies to be useful; they can instead act as temporary, imperfect sources of guidance whose value is tested and internalized through interaction.

Giovanni Bonetta, Matteo Merler, Davide Zago +2

Department of Computer Science, University of Torino, Torino, Italy · Natural Language Processing Unit, Fondazione Bruno Kessler, Trento, Italy

Uncertainty Sampling Visual RL Policy Distillation

Aug 27, 2026cs.RO

SOLO: Stable Omni-terrain Long-Horizon Perceptive Humanoid Locomotion

Humans traverse complex terrain over long distances without losing balance, whereas perceptive humanoid policies become fragile as perception and control errors accumulate. We present SOLO, a unified framework addressing two compounding causes of this long-horizon fragility: dense terrain reconstruction smooths action-critical details, and pointwise imitation lacks temporal credit assignment. Its Query Reconstructor (QR) uses Fourier-encoded cell queries to retrieve spatially specific evidence from depth-proprioception tokens, preserving sharp terrain boundaries. Trajectory-Aware MSE (TA-MSE) Distillation adds next-state teacher-student disagreement to the PPO reward, enabling Generalized Advantage Estimation to propagate future disagreement penalties to preceding actions. In simulation, QR reduces height-map L1 error by factors of 3.3-4.0, while TA-MSE surpasses PPO and MSE+PPO in curriculum progression. On stress-test terrains, SOLO achieves 97.5% mean traversal success and 96% stepping-stone success, versus 75.0-75.6% and 0-3% for dense-reconstructor variants. Deployed zero-shot with only a chest-mounted depth camera and proprioception, SOLO completes a continuous 1.5-km outdoor route and an indoor mixed-terrain course. Project page: https://sunpihai-up.github.io/solo/

Pihai Sun, Gang Han, Jingkai Sun +17

1Artificial General Intelligence Institute, University of Science and Technology of China · 2X-Humanoid · 4The University of Hong Kong +5

Humanoid Robot Locomotion 3D Reconstruction Terrain-Aware Robot Locomotion Robot Locomotion RL for Legged Locomotion Policy Distillation

Aug 6, 2026cs.AI

When History Lies: Evaluating and Improving Tool Use under Misleading Multi-Turn Histories

Tool-calling agents infer task state from accumulated dialogue and tool traces. In persistent interactions, however, historical traces may remain structurally valid and semantically plausible after they cease to be authoritative for the current request. We show that such history can hijack a policy the model already possesses: on Qwen3-1.7B, pollution flips 32.1% of decisions that are correct under the original trajectory and frequently induces reuse of corrupted entities or interface conventions. We introduce bench, a paired benchmark with synchronized Original, Polluted, and Oracle State views that preserve the system policy, current tools, latest request, and gold next action. Eleven gold-preserving interventions isolate failures in decision state, entity binding, and interface execution across complete calls and non-call decisions. We further propose ours, which transfers an Oracle-conditioned teacher policy to a student observing only polluted history through soft supervision on student-generated prefixes. On Qwen3-1.7B, ours achieves 87.0% Balanced Tool-Use Accuracy, outperforming Gold-SFT (66.3%), Oracle sequence distillation (82.3%), and off-policy token distillation (85.0%). The method scales consistently: an 8B teacher raises the same compact 1.7B student to 91.9%, while an 8B student reaches 93.0%. The resulting policies further transfer to clean histories, unseen functions, independently regenerated evaluation contexts, external tool-use benchmarks, and noisy multi-hop question answering. These results establish history reliability as a distinct tool-use bottleneck and demonstrate reliable-state policy transfer as an effective and scalable solution.

Xiaoqing Wu, Xingyu Fan, Feifei Li +1

WeChat, Tencent Inc., Beijing, China

Tool-Use Evaluation LLM Tool Use Policy Distillation LLM Agent Reliability

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

Jul 27, 2026cs.LG

Explainable Reinforcement Learning via Physics-Aware Policy Distillation

In safety-critical sectors such as robotics and automotive engineering, the deployment of Deep Reinforcement Learning (DRL) is often hindered by the black-box nature of deep neural networks. This lack of transparency poses significant challenges for regulatory compliance and human-agent trust. This paper presents an experimental study aimed at making high-performance continuous control DRL systems interpretable. A policy distillation framework is implemented using the classic Inverted Pendulum benchmark. A high-performance Twin Delayed DDPG (TD3) agent serves as an opaque, continuous teacher model, whose policy is distilled into an interpretable student surrogate based on a shallow Decision Tree. By leveraging a custom physics-aware feature and "Noisy Oracle Rollouts" for dataset generation, the distillation process achieves performance equivalent to the expert teacher. Furthermore, comparative control theory analysis reveals a fundamental trade-off: transitioning from continuous to discrete rule-based control induces high-frequency Bang-Bang actuation and a stable bimodal limit cycle. Simulation results indicate that Bounded-Input Bounded-Output (BIBO) stability is maintained while providing both global and local interpretability for safe autonomous systems.

Shaker Al-Tamari, Waled Kadour

Dept. of Electrical and Computer Engineering RPTU Kaiserslautern-Landau Kaiserslautern, Germany

RL Control Interpretability in RL Policy Distillation

Jul 10, 2026cs.RO

CoDiMAD: Diffusion-Based Privileged Distillation for Communication-Free Multi-Robot Coordination

Decentralized multi-robot coordination under partial observability remains challenging, especially in communication-free settings where agents must act solely from local sensor observations. Privileged policy distillation provides a promising approach by transferring knowledge from a globally informed oracle to sensor-constrained students. However, in multi-agent systems, the same local observation may correspond to multiple global configurations requiring qualitatively different cooperative actions, making the conditional action distribution inherently multi-modal. Standard deterministic distillation collapses these modes to their mean, often yielding invalid or hesitant actions. To address this issue, we propose CoDiMAD, a three-stage framework that trains a privileged oracle with MAPPO, constructs an offline dataset of local-observation-oracle-action pairs, and distills the oracle into decentralized students parameterized as conditional denoising diffusion probabilistic models. By approximating the conditional oracle-action distribution through the diffusion reverse process, CoDiMAD samples decisive actions from coherent coordination modes rather than averaging across them. Theoretical analysis characterizes the mode-averaging failure of deterministic distillation and the distributional recovery property of diffusion-based distillation. Experiments on three cooperative tasks show that CoDiMAD consistently outperforms direct local MARL and deterministic distillation baselines. The source code will be made publicly available upon acceptance.

Jiyue Tao, Shunheng Xin, Tongsheng Shen +2

Robotics and Control Laboratory, School of Advanced Manufacturing and Robotics, and the State Key Laboratory of Turbulence and Complex Systems, Peking University, Beijing, 100871, China · National Innovation Institute of Defense Technology, Beijing 100071, China

Multi-Agent Coordination Multi-Robot Coordination Decentralized MARL Policy Distillation Diffusion Models

Jul 6, 2026cs.LG

Trust Region Policy Distillation

Big goals are hard to achieve all at once; breaking them into small steps is wiser. We present Trust Region Policy Distillation (TOP-D), which transforms the notoriously unstable, high-variance On-Policy Distillation (OPD) into a stable training paradigm by dynamically constructing a proximal teacher. Theoretically, we establish a rigorous framework demonstrating that TOP-D inherently controls gradient variance. By providing a formal global convergence analysis alongside a monotonic improvement bound, we mathematically formalize the reliability and stability of the overall training dynamics. Empirically, TOP-D dramatically enhances training stability, sample efficiency, and final performance on mathematical reasoning tasks. More importantly, TOP-D introduces zero additional computational overhead, positioning itself as a promising alternative to the well-established OPD paradigm.

Zhengpeng Xie, Li Lyna Zhang, Zeke Xie +1

On-Policy Distillation Policy Learning Policy Distillation Knowledge Distillation

Jul 3, 2026cs.RO

Exp2VLA: Enabling Vision-Language-Action for Drone Navigation from Expert Demonstrations

Vision-language-action (VLA) models open a new path toward intuitive robot control by directly linking perception, language, and action in a single end-to-end framework. Yet for UAVs, practical adoption remains difficult because existing solutions are either computationally heavy or insufficiently capable in complex environments. In this work, we propose a practical expert-distillation pipeline (Exp2VLA) for language-conditioned drone navigation. The core idea is to distill expert behavior, obtained from reinforcement learning, teleoperation, or other controllers, into training data that can be used to fine-tune compact VLA models. This allows existing control strategies to be transferred into a unified language-guided navigation model, reducing manual system integration and lowering the barrier for deploying new robot behaviors. Experiments in both sim-to-sim and simulation-in-the-loop settings across multi-object scenes show that the fine-tuned models can handle varied semantic commands and generalize to unseen target compositions. The proposed framework demonstrates how expert-policy distillation can help mechatronic systems move from specialized control modules toward more flexible and reusable robot intelligence.

Van Huyen Dang, Kabilesh Rajendran, Erdi Sayar +1

Automatic Control Group Paderborn University Paderborn, Germany

UAV Navigation Vision-Language-Action Models Robot Imitation Learning Aerial VLN Language-Conditioned Robot Control Policy Distillation

Jun 29, 2026quant-ph

Staged Hybridisation for Visual Quantum Reinforcement Learning via Knowledge Distillation

Visual environments are a demanding setting for quantum reinforcement learning (QRL): high-dimensional observations, unstable RL optimisation, and constrained variational quantum circuits (VQCs) are difficult to train jointly. This paper studies knowledge distillation (KD) as a staged hybridisation strategy for visual QRL. Instead of training a hybrid visual agent end-to-end from pixels, we first train a classical visual teacher, freeze its encoder as a feature interface, and distil the teacher's policy behaviour into compact downstream heads. These heads can be classical or VQC-based, enabling small quantum-compatible students to be evaluated under the same frozen representation as compact classical controls. We evaluate the pipeline on CartPole Pixels and Acrobot Pixels. The results show that staged KD enables shallow VQC heads to acquire non-trivial visual-control behaviour in settings where direct pixel-based training would be substantially more difficult. Angle-encoded VQC heads retain near-teacher performance, while amplitude-encoded heads push compactness to an extreme regime, at the cost of greater fragility, stronger budget sensitivity, and higher simulation time. Overall, staged KD reframes visual QRL as a compact-head learning problem, opening a practical route for training small quantum-compatible policies outside the standard end-to-end RL loop.

Javier Lazaro, Juan-Ignacio Vazquez, Pablo Garcia-Bringas

University of Deusto, Bilbao, Spain · FSAS International Quantum Center, Fujitsu

Hybrid Quantum-Classical ML Variational Quantum Circuits Visual RL Policy Distillation Knowledge Distillation

Jun 28, 2026cs.AI

UCOB: Learning to Utilize and Evolve Agentic Skills via Credit-Aware On-Policy Bidirectional Self-Distillation

Skill memories can improve agentic reinforcement learning by reusing past experience as textual guidance, but retrieved skills are not oracular: they may help in one state while misleading the same policy in another. This makes the common privileged-teacher assumption fragile, namely that a skill-conditioned prompt can be treated as a fixed teacher for the no-skill prompt. We introduce UCOB, a framework for learning to utilize and evolve agentic skills via credit-aware on-policy bidirectional self-distillation. UCOB treats skill-conditioned and no-skill prompts as two on-policy context views of the same model, compares their return-to-go within the same task and anchor state, and uses the higher-return view as the local teacher. This local credit signal internalizes useful skill-conditioned behavior, corrects misleading skill usage, and guides task/state skill memory updates, utility-aware retrieval, and reflection self-training. Experiments on agentic tasks, including ALFWorld, WebShop, and Search-QA, show that UCOB outperforms skill-free RL, skill-memory baselines, and self-distillation methods across model scales, with up to 23.5 and 18.0 point gains over SOTA baselines on ALFWorld and WebShop. Ablations and analyses further validate its core mechanisms and efficiency.

Songjun Tu, Chengdong Xu, Qichao Zhang +6

Institute of Automation, Chinese Academy of Sciences · Pengcheng Laboratory · Memorax AI

Credit Assignment in RL LLM Agent Skill Learning Policy Distillation RL for LLM Agents

Jun 9, 2026cs.RO

GuideWalk: Learning Unified Autonomous Navigation and Locomotion for Humanoid Robots across Versatile Terrains

Humanoid robots have achieved strong locomotion capabilities, but reliable navigation on versatile terrains remains challenging because obstacle avoidance must be coordinated with dynamically feasible motion. In this work, we present GuideWalk, a unified end-to-end framework that integrates traversability-aware navigation guidance with terrain-adaptive locomotion teacher for humanoid navigation. Specifically, we introduce a navigation module that provides explicit velocity guidance, decoupling obstacle avoidance from terrain conditions to enable robust planning across diverse environments. We propose a composite teacher distillation scheme, where goal-directed commands and dynamically consistent actions are aggregated and distilled into a single policy. To further improve robustness, the distilled policy is refined with reinforcement learning and an auxiliary behavior cloning objective, which promotes exploration while preserving desirable teacher behaviors. Experiments demonstrate that GuideWalk achieves stable and effective navigation while maintaining stable humanoid locomotion.

Haoxuan Han, Chen Chen, Linao Gong +6

Harbin Institute of Technology · Leju Robotics

Humanoid Robot Locomotion Robot Navigation Terrain-Aware Robot Locomotion Policy Distillation

Jun 7, 2026cs.AI

Distilling LLM Reasoning into an Interpretable Policy Tree for Human-AI Collaboration

Constructing efficient and reliable policies to assist humans is indispensable for human-AI collaboration. Existing methods mainly follow two lines of work. Most prior work relies on multi-agent reinforcement learning (MARL) to learn black-box policies, which limits interpretability and raises safety concerns. Recent methods query large language models (LLMs) at each decision step, causing slow responses and high inference costs. We propose Collaboration Policy Tree (Co-pi-tree), a closed-loop method that learns an executable policy tree consisting of a partner-behavior prediction tree and an agent-action selection tree. Co-pi-tree constructs a policy by distilling LLM reasoning into policy tree code. It then evaluates the policy through partner interaction, obtains feedback, and uses natural language to summarize the interaction feedback to improve problematic branches. Experiments in Overcooked-AI show that Co-pi-tree improves average reward by 35.4% over the baseline average, while reducing the number of LLM queries by 77.7% and test-time latency by 97.1%. Project page: https://beiwenzhang.github.io/Co-pi-tree/

Beiwen Zhang, Yongheng Liang, Guowei Zou +2

Sun Yat-sen University

Reinforcement Learning Interpretability in RL Policy Distillation Human-AI Collaboration

Jun 4, 2026cs.RO

HANDOFF: Humanoid Agentic Task-Space Whole-Body Control via Distilled Complementary Teachers

Humanoid loco-manipulation benefits from one controller to coordinate locomotion, arm movements, and fall recovery. These behaviors are difficult to learn together from scratch because they require different skills and objectives. We present HANDOFF, a training architecture that distills motion-tracking, locomotion, and recovery teachers into one 29-DoF policy, using context-conditioned KL losses over separate action slices and a soft mixture of experts. Commanded velocity continuously blends body supervision, while a binary recovery flag assigns the full action to the recovery teacher. The resulting controller takes a compact 10-D task-space command rather than a dense kinematic stream and does not switch policies at runtime. On the Unitree G1, HANDOFF matches state-of-the-art velocity tracking and obtains the largest robust manipulation workspace in a matched adapted-interface benchmark. The same controller executes natural-language-driven, multi-stage loco-manipulation tasks in simulation and hardware experiments without further data collection or fine-tuning.

Lizhi Yang, Junheng Li, Nehar Poddar +5

California Institute of Technology · The Institute for Human & Machine Cognition

Humanoid Loco-Manipulation Robot Skill Learning Humanoid Robot Control Whole-Body Robotic Manipulation Whole-Body Control Policy Distillation

May 30, 2026cs.LG

Interpretable Policy Distillation for Power Grid Topology Control

Deep reinforcement learning (RL) offers a promising route to real-time power grid operation, yet large neural policies are costly to evaluate, hard to deploy on constrained hardware, and opaque to operators. We ask whether a Proximal Policy Optimization (PPO) agent for grid topology control can be compressed into compact tree-based surrogates without losing operational performance. A PPO teacher is trained on Grid2Op's standard 14-bus environment with a stability-oriented reward, using stress-focused data collection on critical, high-loading states. The policy is then distilled into a decision tree and a random forest. Across held-out validation episodes, both surrogates exceed the teacher in mean reward and survival length at a fraction of the inference cost. The decision tree shows high exact-action agreement with the PPO argmax and near-complete agreement within its top-ranked actions, while remaining small enough to be inspected directly. Feature-importance analysis reveals a representational shift: the PPO policy relies mainly on line-loading signals, while the distilled tree is driven primarily by bus-topology variables. These results suggest that stress-focused distillation can convert a black-box neural controller into a lightweight, auditable rule-like surrogate suited for real-time deployment, while also surfacing risks tied to deterministic actions and topology-specific generalization.

Aleksandra Dmitruka, Karlis Freivalds

University of Latvia, Faculty of Exact Sciences and Technology, Riga, Latvia

RL Control Reinforcement Learning Decision Tree Learning Interpretability in RL Policy Distillation Power Systems

May 28, 2026cs.CL

GAPD: Gold-Action Policy Distillation for Agentic Reinforcement Learning in Knowledge Base Question Answering

Reinforcement learning (RL) is a natural fit for agentic knowledge base question answering (KBQA), where a model must issue executable actions, observe knowledge-base feedback, and eventually return an answer. However, current RL-based KBQA systems mainly optimize sparse rewards from the final answer, leaving intermediate action errors weakly supervised. This is especially limiting for logical-form annotated KBQA benchmarks: gold logical forms can be converted into executable action sequences, but existing pipelines use them mainly for warm-start data construction rather than for on-policy RL updates. We propose GAPD, a training-time Gold-Action Policy Distillation framework that adds dense token-level guidance to outcome-based RL. To align gold actions with on-policy student rollouts, GAPD uses MID-ANCHOR MATCHING: it treats the intermediate entities reached during student exploration and gold execution as state anchors, and matches student states to gold states through these explored entity sets. The current policy conditioned on this aligned gold action serves as a stop-gradient teacher, whose token distribution is distilled back to the ordinary student policy over generated action-token spans. GAPD consistently surpasses the current state of the art on WebQSP, GrailQA, and GraphQ.

Xin Sun, Jianan Xie, Zhongqi Chen +6

University of Science and Technology of China · 2NLPR, MAIS, Institute of Automation, Chinese Academy of Sciences · 3ShanghaiTech University +1

Agentic RL Knowledge Graph Question Answering Policy Distillation Knowledge Distillation

May 14, 2026cs.LG

Critic-Driven Voronoi-Quantization for Distilling Deep RL Policies to Explainable Models

Despite many successful attempts at explaining Deep Reinforcement Learning policies using distillation, it remains difficult to balance the performance-interpretability trade-off and select a fitting surrogate model. In addition to this, traditional distillation only minimizes the distance between the behavior of the original and the surrogate policy while other RL-specific components such as action value are disregarded. To solve this, we introduce a new model-agnostic method called Critic-Driven Voronoi State Partitioning, which partitions a black box control policy into regions where a simple class of model can be optimized using gradient descent. By exploiting the critic value network of the original policy, we iteratively introduce new subpolicies in regions with insufficient value, standing in for a measure of policy complexity. The partitioning, a Voronoi quantizer, uses nearest neighbor lookups to assign a linear function to each point in the state space resulting in a cell-like diagram. We validate our approach on several well known benchmarks and proof that this distillation approaches the original policy using a reasonable sized set of linear functions.

Senne Deproost, Denis Steckelmacher, Ann Nowé

Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, Belgium

Reinforcement Learning Vector Quantization Interpretability in RL Policy Distillation Knowledge Distillation

May 9, 2026cs.AI

Reasoning Compression with Mixed-Policy Distillation

Reasoning-centric large language models (LLMs) achieve strong performance by generating intermediate reasoning trajectories, but often incur excessive token usage and high inference-time decoding cost. We observe that, when solving the same problems, larger reasoning models can often produce more concise traces, whereas smaller reasoning models tend to generate longer and more redundant trajectories. This is especially problematic in real-world deployment, where memory, latency, and serving-cost constraints often favor smaller models. Our observations suggest that reasoning compression can be transferred from large models to small ones rather than enforced through explicit length constraints. Based on this insight, we propose Mixed-Policy Distillation (MPD), a reasoning compression framework that transfers concise reasoning behavior from a larger-sized teacher to a smaller student by distilling teacher-compressed student trajectories. Unlike on-policy distillation, which aligns the student with teacher distributions over verbose student trajectories, or off-policy distillation, which relies on teacher-generated trajectories and may suffer from distribution mismatch, MPD combines the strengths of both. Given a student-sampled trajectory, the teacher rewrites it into a more concise reasoning trace, and the student is trained via KL-based alignment on the compressed trajectory. This preserves student-policy exploration while injecting teacher-guided compression. Experiments on Qwen3-1.7B show that MPD reduces token usage by up to 27.1% while improving performance across multiple reasoning benchmarks, demonstrating an effective approach to efficient small-model reasoning.

Han Yang, Mingyan Wu, Bailan He +4

Technical University of Munich · GESIS – Leibniz Institute for the Social Sciences · Northeastern University +5

Language Model Distillation Efficient Language Model Inference Policy Distillation CoT Compression Large Reasoning Models

Apr 22, 2026cs.CL

Hybrid Policy Distillation for LLMs

Knowledge distillation (KD) is a powerful paradigm for compressing large language models (LLMs), whose effectiveness depends on intertwined choices of divergence direction, optimization strategy, and data regime. We break down the design of existing KD methods and present a unified view that establishes connections between them, reformulating KD as a reweighted log-likelihood objective at the token level. We further propose Hybrid Policy Distillation (HPD), which integrates the complementary advantages of forward and reverse KL to balance mode coverage and mode-seeking, and combines off-policy data with lightweight, approximate on-policy sampling. We validate HPD on long-generation math reasoning as well as short-generation dialogue and code tasks, demonstrating improved optimization stability, computational efficiency, and final performance across diverse model families and scales. The code related to this work is available at https://github.com/zwhong714/Hybrid-Policy-Distillation.

Wenhong Zhu, Ruobing Xie, Rui Wang +1

Department of Computer Science and Engineering, Shanghai Jiao Tong University · Shanghai Innovation Institute · Large Language Department, Tencent

Language Model Distillation On-Policy Distillation Policy Distillation Knowledge Distillation