Imitation Learning

We consider the stabilization of Vlasov--Poisson plasma dynamics, a central control problem in nuclear fusion. Our focus is the gap between what an ideal controller would use and what experiments can actually observe: while optimal policy may rely on the full phase-space state, practical feedback is typically limited to sparse macroscopic diagnostics. We therefore study imitation learning methods that distill a fully observed expert policy into controllers operating only on macroscopic measurements. We show the stability guarantees of the learned policy, where the error floor depends on the minimal behavior cloning loss achievable under the observation constraints. We further characterize this minimal loss in terms of a notion of entropy that quantifies the complexity of the initial distribution. Our results demonstrates the theoretical feasibility of learning stabilizing feedback policies for kinetic plasma dynamics from macroscopic observations, and exhibits the adaptivity of the learning approach to low-complexity structures. Through extensive numerical experiments, we validate our theory and show that the learned policies can stabilize the system using only macroscopic observations, within a significantly longer time horizon than non-adaptive baseline controllers.

High-precision assembly frequently involves tight-tolerance insertions, where even slight pose errors can cause jamming or excessive interaction forces, making robust and safe insertion policies difficult to obtain. This paper proposes a tactile-augmented two-stage method that combines Imitation Learning (IL) and Reinforcement Learning (RL) for precision insertion tasks. In the first stage, IL learns a reaching policy with position generalization that grasps the peg and brings it to the vicinity of the target region. In the second stage, RL executes the insertion and enables recovery from failures during contact-rich interactions. To better exploit tactile feedback, we introduce tactile group sampling to increase coverage of critical contact segments during training, and design a tactile critic to more accurately evaluate policy values, improving insertion performance while maintaining low contact forces. We conduct systematic experiments across five hole geometries and three clearance settings. Results show that our method substantially improves insertion performance across all settings; under the most challenging 0.05,mm clearance, it achieves a 67% success rate while keeping contact forces low, reducing the maximum interaction force by 60% and torque by 44%, thereby validating both effectiveness and safety for precision assembly.

Music-inspired Automatic Stage Lighting Control (ASLC) has gained increasing attention in recent years due to the substantial time and financial costs associated with hiring and training professional lighting engineers. However, existing methods suffer from several notable limitations: the low interpretability of rule-based approaches, the restriction to single-primary-light control in music-to-color-space methods, and the limited transferability of music-to-controlling-parameter frameworks. To address these gaps, we propose SeqLight, a hierarchical deep learning framework that maps music to multi-light Hue-Saturation-Value (HSV) space. Our approach first customizes SkipBART, an end-to-end single primary light generation model, to predict the full light color distribution for each frame, followed by hybrid Imitation Learning (IL) techniques to derive an effective decomposition strategy that distributes the global color distribution among individual lights. Notably, the light decomposition module can be trained under varying venue-specific lighting configurations using only mixed light data and no professional demonstrations, thereby flexibly adapting across diverse venues. In this stage, we formulate the light decomposition task as a Goal-Conditioned Markov Decision Process (GCMDP), construct an expert demonstration set inspired by Hindsight Experience Replay (HER), and introduce a three-phase IL training pipeline, achieving strong generalization capability. To validate our IL solution for the proposed GCMDP, we conduct a series of quantitative analysis and human study. The code and trained models are provided at https://github.com/RS2002/SeqLight .

Most evaluations of autonomous driving policies under adversarial conditions are conducted in simulation, due to cost efficiency and the absence of physical risk. However, purely virtual testing fails to capture structural inconsistencies, supervision constraints, and state-representation effects that arise in real-world data and fundamentally shape policy robustness. This work presents an offline trajectory-learning and adversarial robustness evaluation framework grounded in real-world intersection driving data. Within a controlled data contract, we train and compare three trajectory-learning paradigms: Multi-Layer Perceptron (MLP)-based Behavior Cloning (BC), Transformer-based object-tokenized BC, and inverse reinforcement learning (IRL) formulated within a Generative Adversarial Imitation Learning (GAIL) framework. Models are evaluated using Average Displacement Error (ADE) and Final Displacement Error (FDE). Inference-time robustness is assessed by subjecting trained policies to gradient-based adversarial perturbations across multiple intersection scenarios, yielding a structured robustness evaluation matrix. Results show that state-structure design and architectural inductive biases critically influence adversarial stability, leading to markedly different robustness profiles despite comparable nominal prediction accuracy (ADE < 0.08). Inference-time Projected Gradient Descent (PGD) attacks induce final displacement errors of up to approximately 8 meters. The proposed framework establishes a scalable benchmark for studying offline trajectory learning and adversarial robustness in real-world autonomous driving settings.

Mean Field Games (MFGs) provide a powerful framework for modeling the collective behavior of large populations of interacting agents. In this paper, we address the problem of Imitation Learning (IL) in MFGs subject to common noise, where the population distribution evolves stochastically. This stochasticity compels agents to adopt population-aware policies to respond to aggregate shocks. We formulate two distinct learning objectives: recovering a Nash equilibrium and maximizing performance against an expert population. We investigate two imitation proxies: Behavioral Cloning (BC) and Adversarial (ADV) divergence. We then establish finite-sample error bounds showing that minimizing these proxies effectively controls both the policy's exploitability and its performance gap relative to the expert. Furthermore, we propose a numerical framework using generalized Fictitious Play and Deep Learning to compute expert population-aware policies. Through experiments on three environments we demonstrate that standard population-unaware policies fail to capture the equilibrium dynamics. Our results highlight that learning population-aware policies is crucial to avoid being misled by the randomness inherent in common noise.

Adversarial imitation learning (AIL), a prominent approach in imitation learning, has achieved significant practical success powered by neural network approximation. However, existing theoretical analyses of AIL are primarily confined to simplified settings, such as tabular and linear function approximation, and involve complex algorithmic designs that impede practical implementation. This creates a substantial gap between theory and practice. This paper bridges this gap by exploring the theoretical underpinnings of online AIL with general function approximation. We introduce a novel framework called optimization-based AIL (OPT-AIL), which performs online optimization for reward learning coupled with optimism-regularized optimization for policy learning. Within this framework, we develop two concrete methods: model-free OPT-AIL and model-based OPT-AIL. Our theoretical analysis demonstrates that both variants achieve polynomial expert sample complexity and interaction complexity for learning near-expert policies. To the best of our knowledge, they represent the first provably efficient AIL methods under general function approximation. From a practical standpoint, OPT-AIL requires only the approximate optimization of two objectives, thereby facilitating practical implementation. Empirical studies demonstrate that OPT-AIL outperforms previous state-of-the-art deep AIL methods across several challenging tasks.

Imitation learning (IL) has seen remarkable progress, yet field deployment of IL-powered robots remains hindered by the challenge of out-of-distribution (OOD) scenarios. Fine-tuning pre-trained policies with end-user demonstrations collected in deployment environments is a promising strategy to address this challenge. However, end-user demonstrations are frequently of poor quality, characterized by excessive corrective motions, oscillations, and abrupt adjustments that degrade both learned and fine-tuned policy performance. Existing automated approaches for curating demonstration data require policy rollouts in the environment, making them computationally expensive and impractical for real-world deployment. In this paper, we propose a fast, efficient, and fully automated demonstration ranking metric based on the power spectral density (PSD) of demonstration trajectories. The PSD metric requires no policy learning, environment interaction, or expert labeling, making it well-suited for scalable, in-the-field data curation. Lower PSD values correspond to smoother, higher-quality demonstrations, while higher PSD values indicate erratic, artifact-laden trajectories. We evaluate the proposed metric on two benchmark imitation learning datasets comprising expert and lay-user demonstrations, and through a user study with older adults at a retirement facility, where collected demonstrations are used to fine-tune $\pi0.5$ \cite{intelligence2025pi_} for a daily living task. Results demonstrate that PSD-curated data yields policies with higher task success rates and smoother execution trajectories compared to uncurated baselines and two competitive data-ranking methods.

Imitation learning offers a promising framework for enabling robots to acquire diverse skills from human users. However, most imitation learning algorithms assume access to high-quality demonstrations an unrealistic expectation when collecting data from non-expert users, whose demonstrations often contain inadvertent errors. Naively learning from such demonstrations can result in unsafe policy behavior, while discarding entire demonstrations due to occasional mistakes wastes valuable data, especially in low-data settings. In this work, we introduce GiB (Good-in-Bad), an algorithm that automatically identifies and discards erroneous subtasks within demonstrations while preserving high-quality subtasks. The filtered data can then be used by any policy learning algorithm to train more robust policies. GiB first trains a self-supervised model to learn latent features and assigns binary weights to label each demonstration as good or bad. It then models the latent feature distribution of high-quality segments and uses the Mahalanobis distance to detect and evaluate poor-quality subtasks. We validate GiB on the Franka robot in both simulated and real-world multi-step tasks, demonstrating improved policy performance when learning from mixed-quality human demonstrations.

Recent advances in 3D Gaussian Splatting (3DGS) have enabled visually realistic demonstration generation from a single expert trajectory and a short multi-view scan. However, existing 3DGS-based synthesis pipelines typically generate new motions using sampling-based planners or trajectory optimization, which often deviate substantially from the expert's demonstrated path. While such deviations may be acceptable for tasks insensitive to motion shape, they discard subtle spatial and temporal structure that is critical for contact-rich and shape-sensitive manipulation, causing increased demonstration diversity to harm downstream policy learning. We argue that demonstration synthesis should treat the expert trajectory as a strong prior. Building on this principle, we propose a framework that synthesizes diverse task demonstrations while explicitly preserving expert motion structure. We model the expert trajectory using Dynamic Movement Primitives (DMPs) and retarget it to new goals, object configurations, and viewpoints within a reconstructed 3DGS scene, yielding phase-consistent, shape-preserving motion by construction. To safely realize this expert-preserving diversity in cluttered scenes, we introduce an analytic obstacle-aware DMP formulation that operates directly on the continuous density field induced by the 3DGS representation. This enables collision avoidance while minimally perturbing the nominal expert motion, unifying photorealistic rendering and geometric reasoning without additional scene representations. We evaluate our approach on a Spot mobile manipulator across three manipulation tasks with increasing sensitivity to trajectory fidelity. Compared to planner- and optimization-based synthesis, our method produces trajectories with lower deviation and collision rates and yields higher task success when training diffusion-based visuomotor policies.

Task success has historically been the primary measure of policy performance in imitation learning (IL) research. This characteristics strictly limits the ubiquitous applications of IL algorithms in field robotics where safety assurance, in addition to task-success, is of paramount importance. It is often desirable for an IL-powered robot in the field not to roll out a policy, and hence score a poor performance, if the safety is not guaranteed. Although this trade-off between safety and performance is well investigated in classical control literature, policy safety is a heavily underexplored domain in IL research. There is no universal definition of safety in IL. To make things worst, many existing theoretical works on safety is notoriously difficult to extend to IL-powered robots in the field. This paper offers important insights on the safety and performance of IL policies. We propose execution guarantee, a policy-agnostic safety measure that guarantees the maximum task success for a visuomotor IL policy, despite minor run-time changes, from within a specific region in the state space. We leverage recent advances in view synthesis to identify such regions in the state space for an IL policy and explore a fundamental result on set invariance - namely, Nagumo's sub-tangentiality condition - to prove and operationalize execution guarantee from inside that region. Experiments with a Franka robot, both in simulation and real world, demonstrate how the proposed safety analysis allows various IL policies to achieve maximum task success with guarantee. We also demonstrate some interesting results on how a recovery policy - a by-product of the proposed safety analysis - can help to increase the policy performance and thereby mitigating the safety-performance tradeoff in IL.

Recent imitation learning (IL) algorithms such as flow-matching and diffusion policies demonstrate remarkable performance in learning complex manipulation tasks. However, these policies often fail even when operating within their training distribution due to extreme sensitivity to initial conditions and irreducible approximation errors that lead to compounding drift. This makes it unsafe to deploy IL policies in the field where out-of-distribution scenarios are prevalent. A prerequisite for safe deployment is enabling the policy to determine whether it can execute a task the way it was learned from demonstrations. This paper presents TAIL-Safe, a principled approach to identify, for a trained IL policy, a safe set from where the policy empirically succeeds in completing the learned task. We propose a Lipschitz-continuous Q-value function that maps state-action pairs to a long-term safety score based on three short-term task-agnostic criteria: visibility, recognizability, and graspability. The zero-superlevel set of this function characterizes an empirical control invariant set over state-action pairs. When the nominal policy proposes an action outside this set, we apply a recovery mechanism inspired by Nagumo's theorem that uses gradient ascent to the Q-function to steer the policy back to safety. To learn this Q-function, we construct a high-fidelity digital twin using Gaussian Splatting that enables systematic collection of failure data without risk to physical hardware. Experiments with a Franka Emika robot demonstrate that flow-matching policies, which fail under run-time perturbations, achieve consistent task success when guided by the proposed TAIL-Safe.

End-to-end (E2E) autonomous driving aims to directly map sensory observations to driving actions, but its real-world deployment is hindered by evolving data distributions and the high cost of continual annotation. While combining imitation learning (IL) and reinforcement learning (RL) is a common strategy for policy improvement, conventional RL training relies on delayed, event-based rewards, where policies learn only from catastrophic outcomes such as collisions, leading to premature convergence to suboptimal behaviors. To address these limitations, we propose GSDrive, a framework that uses a differentiable 3D Gaussian Splatting (3DGS) environment for future-aware trajectory probing and reward shaping in E2E driving. GSDrive first learns a multi-mode trajectory probe via IL and then uses RL to evaluate multiple candidate futures in the 3DGS environment, converting their simulated returns into dense shaping rewards for policy optimization. This yields a cyclic hybrid IL-RL training loop, where IL supplies structured future priors and RL provides interactive feedback for iterative refinement. Evaluated on the reconstructed nuScenes dataset, our method outperforms other simulation-based RL approaches in closed-loop experiments. Code is available at https://github.com/ZionGo6/GSDrive.

Motion retargeting from humans to human-like artificial agents is becoming increasingly important as humanoid robots grow more capable. However, most existing approaches focus only on reproducing kinematics and ignore the rich sensorimotor experience associated with human movement. In this work, we present a framework for simulating the multimodal sensorimotor experiences of infants using physical and virtual humanoids. From a single video, our method reconstructs the infant's body configuration by extracting its skeletal structure and estimating the full 3D pose from each frame. Then we map the reconstructed motion onto several developmental platforms: the physical iCub robot and the virtual simulators pyCub, EMFANT and MIMo. Replaying the retargeted motions on these embodiments produces simulated multisensory streams including proprioception (joints and muscles), touch, and vision. For the best-matching embodiment, the retargeting achieves sub-centimeter accuracy and enables a rich multimodal analysis of infant development as well as enhanced automated annotation of behaviors. This framework provides a unique window into the infant's sensorimotor experience, offering new tools for robotics, developmental science, and early detection of neurodevelopmental disorders. The code is available at https://github.com/ctu-vras/motion-retargeting/.

Robotic systems that interact with the physical world must reason about kinematic and dynamic constraints imposed by their own embodiment, their environment, and the task at hand. We introduce KinDER, a benchmark for Kinematic and Dynamic Embodied Reasoning that targets physical reasoning challenges arising in robot learning and planning. KinDER comprises 25 procedurally generated environments, a Gymnasium-compatible Python library with parameterized skills and demonstrations, and a standardized evaluation suite with 13 implemented baselines spanning task and motion planning, imitation learning, reinforcement learning, and foundation-model-based approaches. The environments are designed to isolate five core physical reasoning challenges: basic spatial relations, nonprehensile multi-object manipulation, tool use, combinatorial geometric constraints, and dynamic constraints, disentangled from perception, language understanding, and application-specific complexity. Empirical evaluation shows that existing methods struggle to solve many of the environments, indicating substantial gaps in current approaches to physical reasoning. We additionally include real-to-sim-to-real experiments on a mobile manipulator to assess the correspondence between simulation and real-world physical interaction. KinDER is fully open-sourced and intended to enable systematic comparison across diverse paradigms for advancing physical reasoning in robotics. Website and code: https://prpl-group.com/kinder-site/

Contact-rich manipulation is central to many everyday human activities, requiring continuous adaptation to contact uncertainty and external disturbances through multi-modal perception, particularly vision and tactile feedback. While imitation learning has shown strong potential for learning complex manipulation behaviors, most existing approaches rely on action chunking, which fundamentally limits their ability to react to unforeseen observations during execution. This limitation becomes especially critical in contact-rich scenarios, where physical uncertainty and high-frequency tactile feedback demand rapid, reactive control. To address this challenge, we propose Tube Diffusion Policy (TDP), a novel reactive visual-tactile policy learning framework that bridges diffusion-based imitation learning with tube-based feedback control. By leveraging the expressive power of generative models, TDP learns an observation-conditioned feedback flow around nominal action chunks, forming an action tube that enables fast and adaptive reactions during execution. We evaluate TDP on the widely used Push-T benchmark and three additional challenging visual-tactile dexterous manipulation tasks. Across all benchmarks, TDP consistently outperforms state-of-the-art imitation learning baselines. Two real-world experiments further validate its robust reactivity under contact uncertainty and external disturbances. Moreover, the step-wise correction mechanism enabled by action tube significantly reduces the required denoising steps, making TDP well suited for real-time, high-frequency feedback control in contact-rich manipulation.

Learning from demonstration is widely used for robot navigation, yet it suffers from a fundamental limitation: demonstrations consist predominantly of successful behaviors and provide limited coverage of unsafe states. This limitation leads to poor safety when the robot encounters scenarios beyond the demonstration distribution. Failure experiences, such as collisions, contain essential information about unsafe regions, but remain underutilized. The key difficulty lies in the fact that failure data do not provide valid guidance for action imitation, and their naive incorporation into policy learning often degrades performance. We address this challenge by proposing a failure-aware learning framework that explicitly decouples the roles of success and failure data. In this framework, failure experiences are used to shape value estimation in hazardous regions, while policy learning is restricted to successful demonstrations. This separation enables the effective use of failure data without corrupting policy behavior. We implement this design within an offline reinforcement learning (RL) setting and evaluate it in both simulation and real-world environments. The results show that our framework consistently reduces collision rates while preserving the task success rate, and demonstrate strong generalization across different environments and robot platforms.

Large language model-based agents have recently emerged as powerful approaches for solving dynamic and multi-step tasks. Most existing agents employ planning mechanisms to guide long-term actions in dynamic environments. However, current planning approaches face a fundamental limitation that they operate at a fixed granularity level. Specifically, they either provide excessive detail for simple tasks or insufficient detail for complex ones, failing to achieve an optimal balance between simplicity and complexity. Drawing inspiration from the principle of \textit{progressive refinement} in cognitive science, we propose \textbf{AdaPlan-H}, a self-adaptive hierarchical planning mechanism that mimics human planning strategies. Our method initiates with a coarse-grained macro plan and progressively refines it based on task complexity. It generates self-adaptive hierarchical plans tailored to the varying difficulty levels of different tasks, which can be optimized by imitation learning and capability enhancement. Experimental results demonstrate that our method significantly improves task execution success rates while mitigating overplanning at the planning level, providing a flexible and efficient solution for multi-step complex decision-making tasks. To contribute to the community, our code and data will be made publicly available at https://github.com/import-myself/AHP.

In behavioral cloning (BC), policy performance is fundamentally limited by demonstration data quality. Real-world datasets contain trajectories of varying quality due to operator skill differences, teleoperation artifacts, and procedural inconsistencies, yet standard BC treats all demonstrations equally. Existing curation methods require costly policy training in the loop or manual annotation, limiting scalability. We propose RINSE (Ranking and INdexing Smooth Examples), a lightweight framework for scoring demonstrations based on trajectory smoothness that is policy-architecture-agnostic and operates on trajectory data alone, with TED additionally using a phase-boundary/contact signal. Grounded in motor control theory, which establishes smoothness as a hallmark of skilled movement, RINSE uses two complementary metrics: Spectral Arc Length (SAL), a spectral measure of frequency-domain regularity, and Trajectory-Envelope Distance (TED), a spatial measure of contact-aware geometric deviation. We show that smoothness filtering can reduce the conditional action variance of the retained data distribution, with downstream effects that can be amplified by action chunking and compounding error. On RoboMimic benchmarks, SAL filtering achieves 16% higher success using one-sixth of the data. On real-world manipulation, TED filtering achieves 20% improvement with half the data. As a retrieval-stage filter within STRAP on LIBERO-10, RINSE re-ranking improves mean success by 5.6%. As soft weights in Re-Mix domain reweighting, RINSE scores produce domain allocations highly correlated with the learned Re-Mix allocations (Spearman $ρ\geq 0.89$). These results support smoothness as a useful quality signal across filtering, retrieval, and reweighting settings, especially in noisy or heterogeneous data regimes.

Imitation learning is a well-established approach for machine-learning-based control. However, its applicability depends on having access to demonstrations, which are often expensive to collect and/or suboptimal for solving the task. In this work, we present GCImOpt, an approach to learn efficient goal-conditioned policies by training on datasets generated by trajectory optimization. Our approach for dataset generation is computationally efficient, can generate thousands of optimal trajectories in minutes on a laptop computer, and produces high-quality demonstrations. Further, by means of a data augmentation scheme that treats intermediate states as goals, we are able to increase the training dataset size by an order of magnitude. Using our generated datasets, we train goal-conditioned neural network policies that can control the system towards arbitrary goals. To demonstrate the generality of our approach, we generate datasets and then train policies for various control tasks, namely cart-pole stabilization, planar and three-dimensional quadcopter stabilization, and point reaching using a 6-DoF robot arm. We show that our trained policies can achieve high success rates and near-optimal control profiles, all while being small (less than 80,000 neural network parameters) and fast enough (up to more than 6,000 times faster than a trajectory optimization solver) that they could be deployed onboard resource-constrained controllers. We provide videos, code, datasets and pre-trained policies under a free software license; see our project website https://jongoiko.github.io/gcimopt/.

Standard supervised classification trains models to imitate the exact labels provided by a perfect oracle. This imitation happens in a single pass, restricting the model to a fixed compute budget even when inputs vary in complexity. Moreover, the rigid training objective forces the model to express absolute certainty on its training data, resulting in overconfident predictions during evaluation. We propose Reinforced Iterative Classification (RIC), which replaces the imitative objective with Reinforcement Learning (RL). RIC deploys a recurrent agent that iteratively updates a predictive distribution over classes, receiving reward for stepwise improvement in prediction quality. The value function provides a natural halting criterion by estimating the remaining scope for improvement. We prove that the iterative formulation recovers the same optimal predictions as cross-entropy while yielding an anytime classifier. On image classification benchmarks, RIC matches the accuracy of supervised baselines with improved calibration and learns to allocate computation adaptively across inputs.

Humanoid robots have demonstrated impressive motor skills in a wide range of tasks, yet whole-body control for humanlike long-time, dynamic fighting remains particularly challenging due to the stringent requirements on agility and stability. While imitation learning enables robots to execute human-like fighting skills, existing approaches often rely on switching among multiple single-skill policies or employing a general policy to imitate input reference motions. These strategies suffer from instability when transitioning between skills, as the mismatch of initial and terminal states across skills or reference motions introduces out-of-domain disturbances, resulting in unsmooth or unstable behaviors. In this work, we propose RPG, a hybrid expert policy framework, for smooth and stable humanoid multi-skills transition. Our approach incorporates motion transition randomization and temporal randomization to train a unified policy that generates agile fighting actions with stability and smoothness during skill transitions. Furthermore, we design a control pipeline that integrates walking/running locomotion with fighting skills, allowing humanlike long-time combat of arbitrary duration that can be seamlessly interrupted or transit action policies at any time. Extensive experiments in simulation demonstrate the effectiveness of the proposed framework, and real-world deployment on the Unitree G1 humanoid robot further validates its robustness and applicability.

How might implicit aesthetic perspectives shape what Information Systems (IS) scholarship recognises as worthy of study (or not)? In this hermeneutic literature analysis, we surface foundational aesthetic assumptions underpinning IS research. We identify four perspectives (aesthetics as imitation, sensory experience, world-making, and political doing) that guide how IS scholars perceive and appreciate sociotechnical phenomena. These perspectives influence what becomes recognisable as legitimate research and what remains unseen. By making aesthetic assumptions explicit, we show how they form epistemic infrastructure that conditions horizons of inquiry. We apply this framework to algorithmic management and digitally mediated intimacy, revealing how alternative perspectives open new research questions whilst exposing dimensions that dominant framings overlook. This analysis foregrounds the importance of aesthetic philosophy to IS literature, offering a vocabulary for articulating how aesthetic perspectives shape theorising, method, and contribution.

Recent advances in video generative models enable the synthesis of realistic human-object interaction videos across a wide range of scenarios and object categories, including complex dexterous manipulations that are difficult to capture with motion capture systems. While the rich interaction knowledge embedded in these synthetic videos holds strong potential for motion planning in dexterous robotic manipulation, their limited physical fidelity and purely 2D nature make them difficult to use directly as imitation targets in physics-based character control. We present DeVI (Dexterous Video Imitation), a novel framework that leverages text-conditioned synthetic videos to enable physically plausible dexterous agent control for interacting with unseen target objects. To overcome the imprecision of generative 2D cues, we introduce a hybrid tracking reward that integrates 3D human tracking with robust 2D object tracking. Unlike methods relying on high-quality 3D kinematic demonstrations, DeVI requires only the generated video, enabling zero-shot generalization across diverse objects and interaction types. Extensive experiments demonstrate that DeVI outperforms existing approaches that imitate 3D human-object interaction demonstrations, particularly in modeling dexterous hand-object interactions. We further validate the effectiveness of DeVI in multi-object scenes and text-driven action diversity, showcasing the advantage of using video as an HOI-aware motion planner.

Grounded Multimodal Named Entity Recognition (GMNER) aims to extract named entities and localize their visual regions within image-text pairs, serving as a pivotal capability for various downstream applications. In open-world social media platforms, GMNER remains challenging due to the prevalence of long-tailed, rapidly evolving, and unseen entities. To tackle this, existing approaches typically rely on either external knowledge exploration through heuristic retrieval or internal knowledge exploitation via iterative refinement in Multimodal Large Language Models (MLLMs). However, heuristic retrieval often introduces noisy or conflicting evidence that degrades precision on known entities, while solely internal exploitation is constrained by the knowledge boundaries of MLLMs and prone to hallucinations. To address this, we propose SAKE, an end-to-end agentic framework that harmonizes internal knowledge exploitation and external knowledge exploration via self-aware reasoning and adaptive search tool invocation. We implement this via a two-stage training paradigm. First, we propose Difficulty-aware Search Tag Generation, which quantifies the model's entity-level uncertainty through multiple forward samplings to produce explicit knowledge-gap signals. Based on these signals, we construct SAKE-SeCoT, a high-quality Chain-of-Thought dataset that equips the model with basic self-awareness and tool-use capabilities through supervised fine-tuning. Second, we employ agentic reinforcement learning with a hybrid reward function that penalizes unnecessary retrieval, enabling the model to evolve from rigid search imitation to genuine self-aware decision-making about when retrieval is truly necessary. Extensive experiments on two widely used social media benchmarks demonstrate SAKE's effectiveness.

Trajectory Optimization (TO) solvers exploit known system dynamics to compute locally optimal trajectories through iterative improvements. A downside is that each new problem instance is solved independently; therefore, convergence speed and quality of the solution found depend on the initial trajectory proposed. To improve efficiency, a natural approach is to warm-start TO with initial guesses produced by a learned policy trained on trajectories previously generated by the solver. Diffusion-based policies have recently emerged as expressive imitation learning models, making them promising candidates for this role. Yet, a counterintuitive challenge comes from the local optimality of TO demonstrations: when a policy is rolled out, small non-optimal deviations may push it into situations not represented in the training data, triggering compounding errors over long horizons. In this work, we focus on learning-based warm-starting for gradient-based TO solvers that also provide feedback gains. Exploiting this specificity, we derive a first-order loss for Sobolev learning of diffusion-based policies using both trajectories and feedback gains. Through comprehensive experiments, we demonstrate that the resulting policy avoids compounding errors, and so can learn from very few trajectories to provide initial guesses reducing solving time by $2\times$ to $20 \times$. Incorporating first-order information enables predictions with fewer diffusion steps, reducing inference latency.

Controllable cooperative humanoid manipulation is a fundamental yet challenging problem for embodied intelligence, due to severe data scarcity, complexities in multi-agent coordination, and limited generalization across objects. In this paper, we present SynAgent, a unified framework that enables scalable and physically plausible cooperative manipulation by leveraging Solo-to-Cooperative Agent Synergy to transfer skills from single-agent human-object interaction to multi-agent human-object-human scenarios. To maintain semantic integrity during motion transfer, we introduce an interaction-preserving retargeting method based on an Interact Mesh constructed via Delaunay tetrahedralization, which faithfully maintains spatial relationships among humans and objects. Building upon this refined data, we propose a single-agent pretraining and adaptation paradigm that bootstraps synergistic collaborative behaviors from abundant single-human data through decentralized training and multi-agent PPO. Finally, we develop a trajectory-conditioned generative policy using a conditional VAE, trained via multi-teacher distillation from motion imitation priors to achieve stable and controllable object-level trajectory execution. Extensive experiments demonstrate that SynAgent significantly outperforms existing baselines in both cooperative imitation and trajectory-conditioned control, while generalizing across diverse object geometries. Codes and data will be available after publication. Project Page: http://yw0208.github.io/synagent

In the context of robot learning for manipulation, curated datasets are an important resource for advancing the state of the art; however, available datasets typically only include successful executions or are focused on one particular type of skill. In this short paper, we briefly describe a dataset of various skills performed in the context of coffee preparation. The dataset, which we call COFFAIL, includes both successful and anomalous skill execution episodes collected with a physical robot in a kitchen environment, a couple of which are performed with bimanual manipulation. In addition to describing the data collection setup and the collected data, the paper illustrates the use of the data in COFFAIL to learn a robot policy using imitation learning.

We present SignDPO, a novel multi-level Direct Preference Optimisation (DPO) framework designed to enhance the alignment of skeleton-based Sign Language Translation. While current skeleton-based models have made significant progress using Maximum Likelihood Estimation, they are primarily constrained by an imitation-based paradigm that lacks discriminative sensitivity to the fine-grained spatio-temporal nuances of sign language, often leading to semantic drift. To address this, SignDPO shifts the optimisation goal from simple sequence mimicry to structured preference alignment across spatial, temporal, and linguistic dimensions. Our framework involves three key designs. First, we introduce a hierarchical perturbation strategy to construct spatial and temporal non-preferred samples at both global and local granularities automatically. Second, we propose a self-guiding mechanism that leverages decoder cross-attention scores to identify and perturb semantically salient skeletal regions, forcing the model to distinguish genuine sign signals from structural distortions. Third, we establish an automated language-level preference generator by fine-tuning a dedicated perturbation model, capturing complex output-level failure modes without manual annotation. Extensive experiments on three widely adopted benchmarks, CSL-Daily, How2Sign, and OpenASL, demonstrate that SignDPO consistently outperforms state-of-the-art gloss-free methods and even rivals established gloss-based ones. Our results suggest that multi-level preference alignment is a powerful paradigm for bridging the gap between high-entropy skeletal trajectories and discrete linguistic semantics.

Vision-Language-Action (VLA) models map multimodal inputs directly to robot actions and are typically trained through large-scale imitation learning. While this paradigm has shown strong performance, prevailing VLA training procedures do not explicitly supervise hard physical constraints such as obstacle avoidance or kinematic feasibility. As a result, the geometric structure underlying physically feasible behavior must be inferred only implicitly from demonstrations. In this paper, we study whether introducing explicit feasibility supervision can provide effective structured guidance for VLA policies. We formulate a simple geometry-grounded feasibility objective and integrate it into the training stage of a diffusion-based VLA policy. To evaluate this idea systematically, we use obstacle-aware manipulation as a controlled probe of geometry-dependent physical feasibility. Empirical results show that augmenting VLA training with feasibility supervision improves both physical reliability and overall task performance, while also enhancing learning efficiency in the low-data regime. These findings indicate that explicit feasibility signals can effectively complement imitation-based VLA learning, highlighting their potential for developing more reliable VLA policies.

Whole-body humanoid locomotion is challenging due to high-dimensional control, morphological instability, and the need for real-time adaptation to various terrains using onboard perception. Directly applying reinforcement learning (RL) with reward shaping to humanoid locomotion often leads to lower-body-dominated behaviors, whereas imitation-based RL can learn more coordinated whole-body skills but is typically limited to replaying reference motions without a mechanism to adapt them online from perception for terrain-aware locomotion. To address this gap, we propose a whole-body humanoid locomotion framework that combines skills learned from reference motions with terrain-aware adaptation. We first train a diffusion model on retargeted human motions for real-time prediction of terrain-aware reference motions. Concurrently, we train a whole-body reference tracker with RL using this motion data. To improve robustness under imperfectly generated references, we further fine-tune the tracker with a frozen motion generator in a closed-loop setting. The resulting system supports directional goal-reaching control with terrain-aware whole-body adaptation, and can be deployed on a Unitree G1 humanoid robot with onboard perception and computation. The hardware experiments demonstrate successful traversal over boxes, hurdles, stairs, and mixed terrain combinations. Quantitative results further show the benefits of incorporating online motion generation and fine-tuning the motion tracker for improved generalization and robustness.

Fast execution of contact-rich manipulation is critical for practical deployment, yet providing fast demonstrations for imitation learning (IL) remains challenging: humans cannot demonstrate at high speed, and naively accelerating demonstrations alters contact dynamics and induces large tracking errors. We present a method to autonomously refine time-accelerated demonstrations by repurposing Iterative Reference Learning Control (IRLC) to iteratively update the reference trajectory from observed tracking errors. However, applying IRLC directly at high speed tends to produce larger early-iteration errors and less stable transients. To address this issue, we propose Incremental Iterative Reference Learning Control (I2RLC), which gradually increases the speed while updating the reference, yielding high-fidelity trajectories. We validate on real-robot whiteboard erasing and peg-in-hole tasks using a teleoperation setup with a compliance-controlled follower and a 3D-printed haptic leader. Both IRLC and I2RLC achieve up to 10x faster demonstrations with reduced tracking error; moreover, I2RLC improves spatial similarity to the original trajectories by 22.5% on average over IRLC across three tasks and multiple speeds (3x-10x). We then use the refined trajectories to train IL policies; the resulting policies execute faster than the demonstrations and achieve 100% success rates in the peg-in-hole task at both seen and unseen positions, with I2RLC-trained policies exhibiting lower contact forces than those trained on IRLC-refined demonstrations. These results indicate that gradual speed scheduling coupled with reference adaptation provides a practical path to fast, contact-rich IL.

Imitation learning has enabled robots to acquire complex visuomotor manipulation skills from demonstrations, but deployment failures remain a major obstacle, especially for long-horizon action-chunked policies. Once execution drifts off the demonstration manifold, these policies often continue producing locally plausible actions without recovering from the failure. Existing runtime monitors either require failure data, over-trigger under benign feature drift, or stop at failure detection without providing a recovery mechanism. We present Rewind-IL, a training-free online safeguard framework for generative action-chunked imitation policies. Rewind-IL combines a zero-shot failure detector based on Temporal Inter-chunk Discrepancy Estimate (TIDE), calibrated with split conformal prediction, with a state-respawning mechanism that returns the robot to a semantically verified safe intermediate state. Offline, a vision-language model identifies recovery checkpoints in demonstrations, and the frozen policy encoder is used to construct a compact checkpoint feature database. Online, Rewind-IL monitors self-consistency in overlapping action chunks, tracks similarity to the checkpoint library, and, upon failure, rewinds execution to the latest verified safe state before restarting inference from a clean policy state. Experiments on real-world and simulated long-horizon manipulation tasks, including transfer to flow-matching action-chunked policies, demonstrate that policy-internal consistency coupled with semantically grounded respawning offers a practical route to improved reliability in imitation learning. Supplemental materials are available at https://sjay05.github.io/rewind-il

Large-scale Unmanned Aerial Vehicle (UAV) failures can split an unmanned aerial vehicle swarm network into disconnected sub-networks, making decentralized recovery both urgent and difficult. Centralized recovery methods depend on global topology information and become communication-heavy after severe fragmentation. Decentralized heuristics and multi-agent reinforcement learning methods are easier to deploy, but their performance often degrades when the swarm scale and damage severity vary. We present Physics-informed Graph Adversarial Imitation Learning algorithm (PhyGAIL) that adopts centralized training with decentralized execution. PhyGAIL builds bounded local interaction graphs from heterogeneous observations, and uses physics-informed graph neural network to encode directional local interactions as gated message passing with explicit attraction and repulsion. This gives the policy a physically grounded coordination bias while keeping local observations scale-invariant. It also uses scenario-adaptive imitation learning to improve training under fragmented topologies and variable-length recovery episodes. Our analysis establishes bounded local graph amplification, bounded interaction dynamics, and controlled variance of the terminal success signal. A policy trained on 20-UAV swarms transfers directly to swarms of up to 500 UAVs without fine-tuning, and achieves better performance across reconnection reliability, recovery speed, motion safety, and runtime efficiency than representative baselines.

3D policy learning promises superior generalization and cross-embodiment transfer, but progress has been hindered by training instabilities and severe overfitting, precluding the adoption of powerful 3D perception models. In this work, we systematically diagnose these failures, identifying the omission of 3D data augmentation and the adverse effects of Batch Normalization as primary causes. We propose a new architecture coupling a scalable transformer-based 3D encoder with a diffusion decoder, engineered specifically for stability at scale and designed to leverage large-scale pre-training. Our approach significantly outperforms state-of-the-art 3D baselines on challenging manipulation benchmarks, establishing a new and robust foundation for scalable 3D imitation learning. Project Page: https://r3d-policy.github.io/

We present a novel hierarchical spatiotemporal action tokenizer for in-context imitation learning. We first propose a hierarchical approach, which consists of two successive levels of vector quantization. In particular, the lower level assigns input actions to fine-grained subclusters, while the higher level further maps fine-grained subclusters to clusters. Our hierarchical approach outperforms the non-hierarchical counterpart, while mainly exploiting spatial information by reconstructing input actions. Furthermore, we extend our approach by utilizing both spatial and temporal cues, forming a hierarchical spatiotemporal action tokenizer, namely HiST-AT. Specifically, our hierarchical spatiotemporal approach conducts multi-level clustering, while simultaneously recovering input actions and their associated timestamps. Finally, extensive evaluations on multiple simulation and real robotic manipulation benchmarks show that our approach establishes a new state-of-the-art performance in in-context imitation learning.

Mobile agents powered by vision-language models have demonstrated impressive capabilities in automating mobile tasks, with recent leading models achieving a marked performance leap, e.g., nearly 70% success on AndroidWorld. However, these systems keep their training data closed and remain opaque about their task and trajectory synthesis recipes. We present OpenMobile, an open-source framework that synthesizes high-quality task instructions and agent trajectories, with two key components: (1) The first is a scalable task synthesis pipeline that constructs a global environment memory from exploration, then leverages it to generate diverse and grounded instructions. and (2) a policy-switching strategy for trajectory rollout. By alternating between learner and expert models, it captures essential error-recovery data often missing in standard imitation learning. Agents trained on our data achieve competitive results across three dynamic mobile agent benchmarks: notably, our fine-tuned Qwen2.5-VL and Qwen3-VL reach 51.7% and 64.7% on AndroidWorld, far surpassing existing open-data approaches. Furthermore, we conduct transparent analyses on the overlap between our synthetic instructions and benchmark test sets, and verify that performance gains stem from broad functionality coverage rather than benchmark overfitting. We release data and code at https://njucckevin.github.io/openmobile/ to bridge the data gap and facilitate broader mobile agent research.

Achieving high-fidelity generation in extremely few sampling steps has long been a central goal of generative modeling. Existing approaches largely rely on distillation-based frameworks to compress the original multi-step denoising process into a few-step generator. However, such methods inherently constrain the student to imitate a stronger multi-step teacher, imposing the teacher as an upper bound on student performance. We argue that introducing \textbf{preference alignment awareness} enables the student to optimize toward reward-preferred generation quality, potentially surpassing the teacher instead of being restricted to rigid teacher imitation. To this end, we propose \textbf{Reward-Aware Trajectory Shaping (RATS)}, a lightweight framework for preference-aligned few-step generation. Specifically, teacher and student latent trajectories are aligned at key denoising stages through horizon matching, while a \textbf{reward-aware gate} is introduced to adaptively regulate teacher guidance based on their relative reward performance. Trajectory shaping is strengthened when the teacher achieves higher rewards, and relaxed when the student matches or surpasses the teacher, thereby enabling continued reward-driven improvement. By seamlessly integrating trajectory distillation, reward-aware gating, and preference alignment, RATS effectively transfers preference-relevant knowledge from high-step generators without incurring additional test-time computational overhead. Experimental results demonstrate that RATS substantially improves the efficiency--quality trade-off in few-step visual generation, significantly narrowing the gap between few-step students and stronger multi-step generators.

Test-time compute scaling, the practice of spending extra computation during inference via repeated sampling, search, or extended reasoning, has become a powerful lever for improving large language model performance. Yet deploying these techniques under finite inference budgets requires a decision that current systems largely ignore: which inputs deserve more compute, and which can be answered cheaply? We formalize this as a constrained optimization problem (maximize expected accuracy subject to an average compute budget) and solve it with a two-stage Solve-then-Learn pipeline. In the solve stage, Lagrangian relaxation decomposes the global constraint into per-instance sub-problems, each admitting a closed-form oracle action that optimally prices accuracy against cost. We prove that the induced cost is monotone in the dual variable, enabling exact budget targeting via binary search. In the learn stage, a lightweight classifier is trained to predict oracle actions from cheap input features, amortizing the allocation rule for real-time deployment. We establish that the task-level regret of the learned policy is bounded by its imitation error times the worst-case per-instance gap, yielding a clean reduction from constrained inference to supervised classification. Experiments on MATH and GSM8K with three LLMs (DeepSeek-V3, GPT-4o-mini, Qwen2.5-7B) show that our method consistently outperforms uniform and heuristic allocation baselines, achieving up to 12.8% relative accuracy improvement on MATH under matched budget constraints, while closely tracking the Lagrangian oracle upper bound with over 91% imitation accuracy.

End-to-end autonomous driving models based on Vision-Language-Action (VLA) architectures have shown promising results by learning driving policies through behavior cloning on expert demonstrations. However, imitation learning inherently limits the model to replicating observed behaviors without exploring diverse driving strategies, leaving it brittle in novel or out-of-distribution scenarios. Reinforcement learning (RL) offers a natural remedy by enabling policy exploration beyond the expert distribution. Yet VLA models, typically trained on offline datasets, lack directly observable state transitions, necessitating a learned world model to anticipate action consequences. In this work, we propose a unified understanding-and-generation framework that leverages world modeling to simultaneously enable meaningful exploration and provide dense supervision. Specifically, we augment trajectory prediction with future RGB and depth image generation as dense world modeling objectives, requiring the model to learn fine-grained visual and geometric representations that substantially enrich the planning backbone. Beyond serving as a supervisory signal, the world model further acts as a source of intrinsic reward for policy exploration: its image prediction uncertainty naturally measures a trajectory's novelty relative to the training distribution, where high uncertainty indicates out-of-distribution scenarios that, if safe, represent valuable learning opportunities. We incorporate this exploration signal into a safety-gated reward and optimize the policy via Group Relative Policy Optimization (GRPO). Experiments on the NAVSIM and nuScenes benchmarks demonstrate the effectiveness of our approach, achieving a state-of-the-art PDMS score of 93.7 and an EPDMS of 88.8 on NAVSIM. The code is available at https://zihaosheng.github.io/ExploreVLA/.

We introduce EgoSim, a closed-loop egocentric world simulator that generates spatially consistent interaction videos and persistently updates the underlying 3D scene state for continuous simulation. Existing egocentric simulators either lack explicit 3D grounding, causing structural drift under viewpoint changes, or treat the scene as static, failing to update world states across multi-stage interactions. EgoSim addresses both limitations by modeling 3D scenes as updatable world states. We generate embodiment interactions via a Geometry-action-aware Observation Simulation model, with spatial consistency from an Interaction-aware State Updating module. To overcome the critical data bottleneck posed by the difficulty in acquiring densely aligned scene-interaction training pairs, we design a scalable pipeline that extracts static point clouds, camera trajectories, and embodiment actions from in-the-wild large-scale monocular egocentric videos. We further introduce EgoCap, a capture system that enables low-cost real-world data collection with uncalibrated smartphones. Extensive experiments demonstrate that EgoSim significantly outperforms existing methods in terms of visual quality, spatial consistency, and generalization to complex scenes and in-the-wild dexterous interactions, while supporting cross-embodiment transfer to robotic manipulation. Codes and datasets will be open soon. The project page is at egosimulator.github.io.

Predicting surgical needle trajectories from endoscopic video is critical for robot-assisted suturing, enabling anticipatory planning, real-time guidance, and safer motion execution. Existing methods that directly learn motion distributions from visual observations tend to overlook the sequential dependency among adjacent motion steps. Moreover, sparse waypoint annotations often fail to provide sufficient supervision, further increasing the difficulty of supervised or imitation learning methods. To address these challenges, we formulate image-based needle trajectory prediction as a sequential decision-making problem, in which the needle tip is treated as an agent that moves step by step in pixel space. This formulation naturally captures the continuity of needle motion and enables the explicit modeling of physically plausible pixel-wise state transitions over time. From this perspective, we propose SutureFormer, a goal-conditioned offline reinforcement learning framework that leverages sparse annotations to dense reward signals via cubic spline interpolation, encouraging the policy to exploit limited expert guidance while exploring plausible future motion paths. SutureFormer encodes variable-length clips using an observation encoder to capture both local spatial cues and long-range temporal dynamics, and autoregressively predicts future waypoints through actions composed of discrete directions and continuous magnitudes. To enable stable offline policy optimization from expert demonstrations, we adopt Conservative Q-Learning with Behavioral Cloning regularization. Experiments on a new kidney wound suturing dataset containing 1,158 trajectories from 50 patients show that SutureFormer reduces Average Displacement Error by 58.6% compared with the strongest baseline, demonstrating the effectiveness of modeling needle trajectory prediction as pixel-level sequential action learning.

Imitation learning is a prominent paradigm for robotic manipulation. However, existing visual imitation methods map 2D image observations directly to 3D action outputs, imposing a 2D-3D mismatch that hinders spatial reasoning and degrades robustness. We present VolumeDP, a policy architecture that restores spatial alignment by explicitly reasoning in 3D. VolumeDP first lifts image features into a Volumetric Representation via cross-attention. It then selects task-relevant voxels with a learnable module and converts them into a compact set of spatial tokens, markedly reducing computation while preserving action-critical geometry. Finally, a multi-token decoder conditions on the entire token set to predict actions, thereby avoiding lossy aggregation that collapses multiple spatial tokens into a single descriptor. VolumeDP achieves a state-of-the-art average success rate of 88.8% on the LIBERO simulation benchmark, outperforming the strongest baseline by a substantial 14.8% improvement. It also delivers large performance gains over prior methods on the ManiSkill and LIBERO-Plus benchmarks. Real-world experiments further demonstrate higher success rates and robust generalization to novel spatial layouts, camera viewpoints, and environment backgrounds. Code and videos are available on the project page: https://yzc0731.github.io/VolumeDP/

Motion imitation learning (IL) is increasingly used in robotics and human gait modeling, yet its ability to recover biomechanically consistent joint moments without explicit kinetic information remains unclear. In this study, we examined whether motion imitation alone can estimate reasonable biological joint moments. We compare motion-only IL (MOIL) against a kinetics-aware IL (KAIL) framework that incorporates ground reaction forces (GRF) and center of pressure (CoP) in imitation rewards, with an ablation study to examine the contribution of each kinetic term. Experiments were conducted using walking data from a non-disabled participant at three speeds (0.9, 1.2, and 1.5 m/s). While both MOIL and KAIL achieved comparable kinematic tracking accuracy, MOIL exhibited substantially larger errors in GRF, CoP, and joint moment estimates relative to inverse dynamics references. In contrast, KAIL produced kinetics more consistent with biomechanical values. These findings highlight a fundamental limitation of MOIL approaches, which may lead to erroneous interpretations of gait biomechanics and downstream applications by failing to estimate consistent human-like gait kinetics.

The acquisition of large-scale physical interaction data, a critical prerequisite for modern robot learning, is severely bottlenecked by the prohibitive cost and scalability limits of human-in-the-loop collection paradigms. To break this barrier, we introduce Robust Autonomous Data Acquisition for Robotics (RADAR), a fully autonomous, closed-loop data generation engine that completely removes human intervention from the collection cycle. RADAR elegantly divides the cognitive load into a four-module pipeline. Anchored by 2-5 3D human demonstrations as geometric priors, a Vision-Language Model first orchestrates scene-relevant task generation via precise semantic object grounding and skill retrieval. Next, a Graph Neural Network policy translates these subtasks into physical actions via in-context imitation learning. Following execution, the VLM performs automated success evaluation using a structured Visual Question Answering pipeline. Finally, to shatter the bottleneck of manual resets, a Finite State Machine orchestrates an autonomous environment reset and asymmetric data routing mechanism. Driven by simultaneous forward-reverse planning with a strict Last-In, First-Out causal sequence, the system seamlessly restores unstructured workspaces and robustly recovers from execution failures. This continuous brain-cerebellum synergy transforms data collection into a self-sustaining process. Extensive evaluations highlight RADAR's exceptional versatility. In simulation, our framework achieves up to 90% success rates on complex, long-horizon tasks, effortlessly solving challenges where traditional baselines plummet to near-zero performance. In real-world deployments, the system reliably executes diverse, contact-rich skills (e.g., deformable object manipulation) via few-shot adaptation without domain-specific fine-tuning, providing a highly scalable paradigm for robotic data acquisition.

Generalizing skill policies to novel conditions remains a key challenge in robot learning. Imitation learning methods, while data-efficient, are largely confined to the training region and consistently fail on input data outside it, leading to unpredictable policy failures. Alternatively, transfer learning approaches offer methods for trajectory generation robust to both changes in environment and tasks, but they remain data-hungry and lack accuracy in zero-shot generalization. We address these challenges in the context of task inversion learning and propose a novel joint learning approach to achieve accurate and efficient knowledge transfer. Our method constructs a common representation of the forward and inverse tasks, and leverages auxiliary forward demonstrations from novel configurations to successfully execute the corresponding inverse tasks, without any direct supervision. We demonstrate the extrapolation capabilities of our framework through ablation studies and experiments in simulated and real-world environments that require complex manipulation skills with a diverse set of objects and tools, where we outperform diffusion-based and multimodal VAE alternatives.

Post-training with reinforcement learning (RL) has recently shown strong promise for advancing multimodal agents beyond supervised imitation. However, RL remains limited by poor data efficiency, particularly in settings where interaction data are scarce and quickly become outdated. To address this challenge, GIPO (Gaussian Importance sampling Policy Optimization) is proposed as a policy optimization objective based on truncated importance sampling, replacing hard clipping with a log-ratio-based Gaussian trust weight to softly damp extreme importance ratios while maintaining non-zero gradients. Theoretical analysis shows that GIPO introduces an implicit, tunable constraint on the update magnitude, while concentration bounds guarantee robustness and stability under finite-sample estimation. Experimental results show that GIPO achieves state-of-the-art performance among clipping-based baselines across a wide range of replay buffer sizes, from near on-policy to highly stale data, while exhibiting superior bias--variance trade-off, high training stability and improved sample efficiency. Code is available at https://github.com/distanceLu/GIPO.

In this work, we present the first theoretical analysis of multi-agent imitation learning (MAIL) in linear Markov games where both the transition dynamics and each agent's reward function are linear in some given features. We demonstrate that by leveraging this structure, it is possible to replace the state-action level "all policy deviation concentrability coefficient" (Freihaut et al., arXiv:2510.09325) with a concentrability coefficient defined at the feature level which can be much smaller than the state-action analog when the features are informative about states' similarity. Furthermore, to circumvent the need for any concentrability coefficient, we turn to the interactive setting. We provide the first, computationally efficient, interactive MAIL algorithm for linear Markov games and show that its sample complexity depends only on the dimension of the feature map $d$. Building on these theoretical findings, we propose a deep MAIL interactive algorithm which clearly outperforms BC on games such as Tic-Tac-Toe and Connect4.

LLM-powered AI assistants acting on behalf of users can produce poor collective outcomes at scale. We introduce a framework for evaluating their emergent behaviour in social dilemmas, applied to three iterated games (Public Goods, Collective Risk, Common Pool Resource). We prompt each model to produce a natural-language strategy, then have the same model translate it into code. This aims to isolate strategic reasoning from input-parsing, enables pre-deployment inspection, and scales to populations of hundreds of agents. We propose three analyses: behavioural fingerprinting via exhaustive evaluation over opponent histories; self-play robustness across mixtures of a model's strategies with either a Selfish or Collective disposition; and cultural evolution under payoff-biased imitation. Applied to three state-of-the-art LLMs, we find substantial cross-model differences in self-play welfare, and that cultural evolution converges to low-welfare, Selfish-dominant equilibria in larger groups.

Reinforcement learning (RL) promises to unlock capabilities beyond imitation learning for Vision--Language--Action (VLA) models, but its requirement for massive real-world interaction prevents direct deployment on physical robots. Recent work attempts to use learned world models as simulators for policy optimization, yet closed-loop imagined rollouts inevitably suffer from hallucination and long-horizon error accumulation. Such errors not only degrade visual fidelity, but also mislead policy optimization by providing unreliable learning signals. We propose WoVR, a reliable world-model-based RL framework for post-training VLA policies. Instead of assuming a faithful world model, WoVR explicitly regulates how RL interacts with imperfect imagined dynamics. It improves rollout stability through a controllable action-conditioned video world model, reshapes imagined interaction to reduce effective error depth via Keyframe-Initialized Rollouts, and maintains policy--simulator alignment through World Model-Policy co-evolution. Extensive experiments demonstrate that WoVR enables stable long-horizon imagined rollouts and effective policy optimization, achieving superior LIBERO performance and consistent real-world gains across multiple robotic platforms. These results show that world models can serve as practical simulators for RL when hallucination is explicitly controlled. Additional visualization results are available at https://wovr-corl.github.io.

Post-training of flow matching models-aligning the output distribution with a high-quality target-is mathematically equivalent to imitation learning. While Supervised Fine-Tuning mimics expert demonstrations effectively, it cannot correct policy drift in unseen states. Preference optimization methods address this but require costly preference pairs or reward modeling. We propose Flow Matching Adversarial Imitation Learning (FAIL), which minimizes policy-expert divergence through adversarial training without explicit rewards or pairwise comparisons. We derive two algorithms: FAIL-PD exploits differentiable ODE solvers for low-variance pathwise gradients, while FAIL-PG provides a black-box alternative for discrete or computationally constrained settings. Fine-tuning FLUX with only 13,000 demonstrations from Nano Banana pro, FAIL achieves competitive performance on prompt following and aesthetic benchmarks. Furthermore, the framework generalizes effectively to discrete image and video generation, and functions as a robust regularizer to mitigate reward hacking in reward-based optimization. Code and data are available at https://github.com/HansPolo113/FAIL.

Contact-rich teleoperation logs expose a policy I/O design choice: demonstrations may contain the robot execution (E), leader command (C), or both. These signals are not interchangeable: E2E may discard contact-generating command offsets, whereas E2C preserves these offsets but omits the robot's execution response. We propose Dual-State Conditioning (EC2C), which conditions on both E and C while predicting future C, exposing command-execution mismatch as a cue for contact, latency, payload, and operator compensation; in quasi-static contact, this cue is often force-correlated. On a low-cost setup without force, tactile, or motor-current policy input, EC2C outperforms E2E and a strong E2C baseline across several real-world contact-rich, force-sensitive, and dynamic tasks. These results support EC2C as a practical default I/O setting for contact-rich imitation learning. We further formulate latency-adaptive inpainting as a temporal extension of this I/O choice for action-chunking policies, and discuss when long histories help dynamic inference or introduce causal confounding.

Realizing versatile and human-like performance in high-demand sports like badminton remains a formidable challenge for humanoid robotics. Unlike standard locomotion or static manipulation, this task demands a seamless integration of explosive whole-body coordination and precise, timing-critical interception. While recent advances have achieved lifelike motion mimicry, bridging the gap between kinematic imitation and functional, physics-aware striking without compromising stylistic naturalness is non-trivial. To address this, we propose Imitation-to-Interaction, a progressive reinforcement learning framework designed to evolve a robot from a "mimic" to a capable "striker." Our approach establishes a robust motor prior from human data, distills it into a compact, model-based state representation, and stabilizes dynamics via adversarial priors. Crucially, to overcome the sparsity of expert demonstrations, we introduce a manifold expansion strategy that generalizes discrete strike points into a dense interaction volume. We validate our framework through the mastery of diverse skills, including lifts and drop shots, in simulation. Furthermore, we demonstrate the first zero-shot sim-to-real transfer of anthropomorphic badminton skills to a humanoid robot, successfully replicating the kinetic elegance and functional precision of human athletes in the physical world.

Semi-supervised imitation learning (SSIL) consists in learning a policy from a small dataset of action-labeled trajectories and a much larger dataset of action-free trajectories. Some SSIL methods learn an inverse dynamics model (IDM) to predict the action from the current state and the next state. An IDM can act as a policy when paired with a video model (VM-IDM) or as a label generator to perform behavior cloning on action-free data (IDM labeling). In this work, we first show that VM-IDM and IDM labeling learn the same policy in a limit case, which we call the IDM-based policy. We then argue that the previously observed advantage of IDM-based policies over behavior cloning is due to the superior sample efficiency of IDM learning, which we attribute to two causes: (i) the ground-truth IDM tends to be contained in a lower complexity hypothesis class relative to the expert policy, and (ii) the ground-truth IDM is often less stochastic than the expert policy. We argue these claims based on insights from statistical learning theory and novel experiments, including a study of IDM-based policies using recent architectures for unified video-action prediction (UVA). Motivated by these insights, we finally propose an improved version of the existing LAPO algorithm for latent action policy learning. We experiment on the Procgen, Push-T and LIBERO benchmarks.

Learning long-horizon robotic manipulation requires jointly achieving expressive behavior modeling, real-time inference, and stable execution, which remains challenging for existing generative policies. Diffusion-based approaches offer strong modeling capacity but incur high inference latency, while flow matching enables fast, near-single-step generation yet often suffers from unstable execution when operating directly in the raw action space. We propose Continuous Latent Action Flow Policy (CoLA-Flow Policy), a trajectory-level imitation learning framework that performs flow matching in a continuous latent action space. By encoding action sequences into temporally coherent latent trajectories and learning an explicit latent-space flow, CoLA-Flow Policy decouples global motion structure from low-level control noise, enabling smooth and reliable long-horizon execution. The framework further integrates geometry-aware point cloud conditioning and execution-time multimodal modulation, using visual cues as a representative modality to enhance real-world robustness. Experiments in simulation and on real robots show that CoLA-Flow Policy achieves near-single-step inference, improves trajectory smoothness by up to 93.7% and task success by up to 25 percentage points over raw action-space flow baselines, while remaining significantly faster than diffusion-based policies.

Behavior cloning (BC) is a practical offline imitation learning method, but it often fails when expert demonstrations are limited. Recent works have introduced a class of architectures named predictive inverse dynamics models (PIDMs) that combine a future-state predictor with an inverse dynamics model. While PIDMs often outperform BC, the reasons behind their benefits remain unclear. In this paper, we provide a theoretical explanation: PIDMs introduce a tradeoff. Conditioning the IDM on the predicted future state can significantly reduce variance, but the prediction itself introduces additional bias and variance. We establish conditions for PIDMs to achieve higher sample efficiency and lower prediction error than BC, with the gap widening when additional data sources are available. We validate the theoretical insights empirically in 2D navigation tasks, where BC requires up to five times (three times on average) more demonstrations than PIDM to reach comparable performance. Results are also illustrated in a complex 3D environment in a modern video game with high-dimensional visual inputs and stochastic transitions, where BC requires over 66% more samples than PIDM.

Scaling imitation learning to diverse multi-task robot manipulation remains challenging due to suboptimal demonstrations, behavioral multi-modality, and destructive interference across tasks. While skill-based methods offer a promising direction by decomposing behaviors into reusable abstractions, existing approaches often learn skills that are either biased toward linguistic structure or lack semantic alignment across tasks, limiting generalization. In this work, we propose AtomSkill, a novel framework that learns a semantically aligned Atomic Skill Space from demonstrations and enables robust long-horizon execution through keypose imagination. Our method introduces: (1) semantic contrastive skill alignment, which partitions demonstrations into variable-length atomic skills and employs a contrastive objective to jointly enforce semantic consistency and temporal coherence, yielding a compact and reusable skill library; and (2) action decoding with keypose imagining, where the policy predicts both a skill's terminal keypose and immediate actions, thereby supporting progress-aware skill transitions. During inference, an atomic skill diffusion sampler generates plausible skill sequences, while predicted keyposes autonomously trigger smooth skill chaining. Extensive experiments in simulation and real-world settings show that AtomSkill consistently outperforms state-of-the-art imitation learning and skill-based baselines. Project page: https://atom-skill.github.io.

Vision-based imitation learning has enabled impressive robotic manipulation skills, but action imitation alone provides limited supervision of the geometric consequences of robot behavior. To address this limitation, we introduce Implicit Scene Supervision (ISS) Policy, a 3D visuomotor diffusion policy with a DiT backbone that predicts continuous action sequences from point-cloud observations. ISS augments action diffusion with a supervised robot motion predictor that maps generated actions and robot-state context to end-effector motion, and then uses the predicted motion together with gripper intent to forecast future point-cloud representations. By explicitly modeling the intermediate transition from action to robot motion, ISS encourages the policy to capture how its actions affect the surrounding 3D scene. We further introduce asymmetric gradient routing to separate direct motion regression from scene-level policy supervision, together with a change-balanced objective that accounts for variations in scene-change magnitude. These auxiliary objectives provide dynamics-aware geometric supervision using only expert demonstrations, without requiring additional annotations or auxiliary modules at inference time. ISS Policy achieves state-of-the-art performance on single-arm manipulation tasks in MetaWorld and dexterous manipulation tasks in Adroit, while real-world dual-arm experiments further demonstrate its effectiveness on physical robotic manipulation. The resulting framework preserves the scalable DiT backbone and standard diffusion-policy control interface. Code and videos will be released.

The collection of large-scale and diverse robot demonstrations remains a major bottleneck for imitation learning, as real-world data acquisition is costly and simulators offer limited diversity and fidelity with pronounced sim-to-real gaps. While generative models present an attractive solution, existing methods often alter only visual appearances without creating new behaviors, or suffer from embodiment inconsistencies that yield implausible motions. To address these limitations, we introduce AnchorDream, an embodiment-aware world model that repurposes pretrained video diffusion models for robot data synthesis. AnchorDream conditions the diffusion process on robot motion renderings, anchoring the embodiment to prevent hallucination while synthesizing objects and environments consistent with the robot's kinematics. Starting from only a handful of human teleoperation demonstrations, our method scales them into large, diverse, high-quality datasets without requiring explicit environment modeling. Experiments show that the generated data leads to consistent improvements in downstream policy learning, with relative gains of 36.4% in simulator benchmarks and nearly double performance in real-world studies. These results suggest that grounding generative world models in robot motion provides a practical path toward scaling imitation learning.

Dexterous robotic hands are expected to perform complex, contact-rich object manipulation, but learning such skills remains challenging because high-dimensional hands require high-fidelity demonstrations. Imitation learning provides a practical route for acquiring dexterous manipulation skills from human demonstrations, yet collecting synchronized multimodal demonstrations with accurate hand actions and tactile observations remains a key bottleneck. We present MILE, a teleoperation-based data-collection system comprising the human-first MILE exoskeleton and the mechanically corresponding MILE-Tac robotic hand. The system integrates custom-designed and fabricated modular joint encoders and compact MILE fingertip visuotactile sensor modules. The exoskeleton is informed by human-hand anatomy and ergonomic constraints, while the robotic hand is co-designed to preserve the selected four-finger kinematic topology. This correspondence enables joint-space command transfer and reduces reliance on task-space IK-based retargeting. The system synchronously records task-specific visual observations, four fingertip visuotactile streams, robot-hand proprioception, and exoskeleton-derived action commands. We evaluate MILE through a four-task teleoperation benchmark against representative glove-based and vision-based interfaces, and through imitation-learning experiments that compare policies trained with and without fingertip tactile input. The project page is available at https://sites.google.com/view/mile-system.

Current navigation benchmarks focus on task success but do not capture the economic constraints essential for commercializing autonomous delivery systems. We introduce CostNav, an Economic Navigation Benchmark that evaluates physical AI agents on a cost-revenue and break-even analysis, pairing Isaac Sim's collision and cargo dynamics with industry-standard data such as Securities and Exchange Commission (SEC) filings and Abbreviated Injury Scale (AIS) injury reports. To our knowledge, CostNav is the first physics-grounded economic benchmark to use regulatory and financial data to quantify the gap between navigation metrics and commercial deployment, revealing that high task-success rates alone do not ensure economic viability. Evaluating seven baselines (two rule-based and five imitation-learning methods), we find no method economically viable: all yield negative contribution margins. CANVAS, using only an RGB camera and GPS, attains the highest task success and the least-negative margin among methods with non-zero Service-Level Agreement (SLA) compliance (-$28.40/run), outperforming LiDAR-equipped Nav2 w/ GPS (-$37.34/run). A sim-trained policy evaluated on a real delivery robot yields SLA compliance close to its simulation result, indicating that policy performance in CostNav's simulation transfers to real-world deployment. We challenge the community to achieve economic viability on CostNav, which scores methods by cost-revenue outcomes. All resources are available at https://github.com/worv-ai/CostNav.