Humanoid robots are expected to operate reliably over long horizons while executing versatile whole-body skills. Yet Reinforcement Learning (RL) motion policies typically lose stability under prolonged operation, sensor/actuator noise, and real world disturbances. In this work, we propose a Selective Adversarial Attack for Robust Training (SA2RT) to enhance the robustness of motion skills. The adversary is learned to identify and sparsely perturb the most vulnerable states and actions under an attack-budget constraint, thereby exposing true weakness without inducing conservative overfitting. The resulting non-zero sum, alternating optimization continually strengthens the motion policy against the strongest discovered attacks. We validate our approach on the Unitree G1 humanoid robot across perceptive locomotion and whole-body control tasks. Experimental results show that adversarially trained policies improve the terrain traversal success rate by 40%, reduce the trajectory tracking error by 32%, and maintain long horizon mobility and tracking performance. Together, these results demonstrate that selective adversarial attacks are an effective driver for learning robust, long horizon humanoid motion skills.
论文检索
输入标题、作者或关键词，从 2,942 篇学术成果中精准定位
Vision-Language-Action (VLA) models frequently encounter challenges in generalizing to real-world environments due to inherent discrepancies between observation and action spaces. Although training data are collected from diverse camera perspectives, the models typically predict end-effector poses within the robot base coordinate frame, resulting in spatial inconsistencies. To mitigate this limitation, we introduce the Observation-Centric VLA (OC-VLA) framework, which grounds action predictions directly in the camera observation space. Leveraging the camera’s extrinsic calibration matrix, OC-VLA transforms end-effector poses from the robot base coordinate system into the camera coordinate system, thereby unifying prediction targets across heterogeneous viewpoints. This lightweight, plug-and-play strategy ensures robust alignment between perception and action, substantially improving model resilience to camera viewpoint variations. The proposed approach is readily compatible with existing VLA architectures, requiring no substantial modifications. Comprehensive evaluations on both simulated and real-world robotic manipulation tasks demonstrate that OC-VLA accelerates convergence, enhances task success rates, and improves cross-view generalization.
MoLe-VLA: Dynamic Layer-skipping Vision Language Action Model via Mixture-of-Layers for Efficient Robot Manipulation
PDF ↗Vision-Language-Action (VLA) models enable robotic systems to perform embodied tasks but face deployment challenges due to the high computational demands of the dense Large Language Models (LLMs), with existing early-exit-based sparsification methods often overlooking the critical semantic role of final layers in downstream tasks. Aligning with the recent breakthrough of the Shallow Brain Hypothesis (SBH) in neuroscience and the mixture of experts in model sparsification, we conceptualize each LLM layer as an expert and propose a Mixture-of-LayEr Vision Language Action model (MoLe-VLA or simply MoLe) architecture for dynamic LLM layer activation. Specifically, we introduce a Spatial-Temporal Aware Router (STAR) for MoLe to selectively activate only parts of the layers based on the robot’s current state, mimicking the brain's distinct signal pathways specialized for cognition and causal reasoning. Additionally, to compensate for the cognition ability of LLM lost during the layer-skipping, we devise a Cognitive self-Knowledge Distillation (CogKD) to enhance the understanding of task demands and generate task-relevant action sequences by leveraging cognition features. Extensive experiments in RLBench simulations and real-world environments demonstrate the superiority of MoLe-VLA in both efficiency and performance, improving the mean success rate by 9.7% across ten simulation tasks while accelerating inference by 36.8% over OpenVLA.
Vision-Language-Action (VLA) models based on flow matching have shown excellent performance in general-purpose robotic manipulation tasks. However, the action accuracy of these models on complex downstream tasks is unsatisfactory. One important reason is that these models rely solely on the post-training paradigm of imitation learning, which makes it difficult to have a deeper understanding of the distribution properties of data quality, which is exactly what Reinforcement Learning (RL) excels at. In this paper, we theoretically propose an offline RL post-training objective for VLA flow models and induce an efficient and feasible offline RL fine-tuning algorithm −− Adaptive Reinforced Flow Matching (ARFM). By introducing an adaptively adjusted scaling factor in the VLA flow model loss, we construct a principled bias-variance trade-off objective function to optimally control the impact of RL signal on flow loss. ARFM adaptively balances RL advantage preservation and flow loss gradient variance control, resulting in a more stable and efficient fine-tuning process. Extensive simulation and real-world experimental results show that ARFM exhibits excellent generalization, robustness, few-shot learning, and continuous learning performance.
Existing vision-and-language navigation models often deviate from the correct trajectory when executing instructions. However, these models lack effective error correction capability, hindering their recovery from errors. To address this challenge, we propose Self-correction Flywheel, a novel post-training paradigm. Instead of considering the model’s error trajectories on the training set as a drawback, our paradigm emphasizes their significance as a valuable data source. We have developed a method to identify deviations in these error trajectories and devised innovative techniques to automatically generate self-correction data for perception and action. These self-correction data serve as fuel to power the model’s continued training. The brilliance of our paradigm is revealed when we re-evaluate the model on the training set, uncovering new error trajectories. At this time, the self-correction flywheel begins to spin. Through multiple flywheel iterations, we progressively enhance our monocular RGB-based VLA navigation model CorrectNav. Experiments on R2R-CE and RxR-CE benchmarks show CorrectNav achieves new state-of-the-art success rates of 65.1% and 69.3%, surpassing prior best VLA navigation models by 8.2% and 16.4%. Real robot tests in various indoor and outdoor environments demonstrate \method's superior capability of error correction, dynamic obstacle avoidance, and long instruction following.
Robot navigation in dense crowds requires understanding social cues that humans naturally use, yet existing methods struggle with real-world complexity. We investigate two questions: (1) Where do pedestrians look when navigating crowds? and (2) Can eye tracking improve robot navigation? To answer, we introduce GazeNav, an egocentric dataset collected via wearable eye trackers, featuring synchronized video, gaze, and trajectories in crowded environments. Analysis reveals that the gaze of pedestrians is closely related to the semantic presence and movement of other individuals, exhibiting distinct attention patterns across navigation behaviors. Building on this, we propose Gaze2Nav, a modular framework that first predicts human gaze to infer socially salient pedestrians, then incorporates the semantic attention into motion planning alongside visual inputs. Our method achieves 87.6% salient pedestrian prediction accuracy and reduces trajectory error by 15.4% over state-of-the-art baselines. By aligning with human gaze, our framework improves both performance and interpretability, advancing toward human-like, socially intelligent robot navigation.
GraphGrasp: Lightweight and Efficient Graph-Guided 6-DoF Robotic Grasp Pose Estimation Network
PDF ↗6-DoF object grasping is a crucial skill for embodied intelligent robots. Previous methods often rely on large-scale networks for feature extraction, followed by grasp pose prediction, which increases the network's parameter count and overlooks the geometric and graph features of the point cloud. To address these challenges, we propose GraphGrasp, a graph-guided 6-DoF grasping pose prediction method. It performs graph analysis from the perspectives of scene, object, and grasping graphs. First, we introduce a graph feature embedding method based on local-global features to model the scene graph effectively. Then, we use a graph transformer strategy to represent spatial relationships between objects in the object graph. Finally, we propose a multi-metric, multi-level grasp pose evaluation algorithm to predict and explore graspable points, enabling effective construction of grasp graphs and accurate grasp pose evaluation. We test GraphGrasp on the GraspNet-1Billion dataset, and the results show that, compared to previous methods, it achieves nearly the same performance with about 1/5 of the parameters of state-of-the-art methods, significantly improving grasp pose prediction speed. Additionally, in real-world robot grasping scenarios, GraphGrasp outperforms previous methods in practical grasp pose prediction tasks.
Actor-Critic for Continuous Action Chunks: A Reinforcement Learning Framework for Long-Horizon Robotic Manipulation with Sparse Reward
PDF ↗Existing reinforcement learning (RL) methods struggle with long-horizon robotic manipulation tasks, particularly those involving sparse rewards. While action chunking is a promising paradigm for robotic manipulation, using RL to directly learn continuous action chunks in a stable and data-efficient manner remains a critical challenge. This paper introduces AC3 (Actor-Critic for Continuous Chunks), a novel RL framework that learns to generate high-dimensional, continuous action sequences. To make this learning process stable and data-efficient, AC3 incorporates targeted stabilization mechanisms for both the actor and the critic. First, to ensure reliable policy improvement, the actor is trained with an asymmetric update rule, learning exclusively from successful trajectories. Second, to enable effective value learning despite sparse rewards, the critic's update is stabilized using intra-chunk n-step returns and further enriched by a self-supervised module providing intrinsic rewards at anchor points aligned with each action chunk. We conducted extensive experiments on 25 tasks from the BiGym and RLBench benchmarks. Results show that by using only a few demonstrations and a simple model architecture, AC3 achieves superior success rates on most tasks, validating its effective design.
Zero-Shot Robotic Manipulation via 3D Gaussian Splatting-Enhanced Multimodal Retrieval-Augmented Generation
PDF ↗Existing end-to-end approaches of robotic manipulation often lack generalization to unseen objects or tasks due to limited data and poor interpretability. While recent Multimodal Large Language Models (MLLMs) demonstrate strong commonsense reasoning, they struggle with geometric and spatial understanding required for pose prediction. In this paper, we propose RobMRAG, a 3D Gaussian Splatting-Enhanced Multimodal Retrieval-Augmented Generation (MRAG) framework for zero-shot robotic manipulation. Specifically, We construct a multi-source manipulation knowledge base containing object contact frames, task completion frames, and pose parameters. During inference, a Hierarchical Multimodal Retrieval module first employs hybrid semantic search to find task-relevant object prototypes, then selects the geometrically closest reference example based on pixel-level similarity and Instance Matching Distance (IMD). We further introduce a 3D-Aware Pose Refinement module based on 3D Gaussian Splatting into the MRAG framework, which aligns the pose of the reference object to the target object in 3D space. The aligned results are reprojected onto the image plane and used as input to the MLLM to enhance the generation of the final pose parameters. Extensive experiments show that on a test set containing 30 categories of household objects, our method improves the success rate by 7.76% compared to the best-performing zero-shot baseline under the same setting, and by 6.54% compared to the state-of-the-art supervised baseline. Our results validate that RobMRAG effectively bridges the gap between high-level semantic reasoning and low-level geometric execution, enabling robotic systems that generalize to unseen objects while remaining inherently interpretable.
ForeDiffusion: Foresight-Conditioned Diffusion Policy via Future View Construction for Robot Manipulation
PDF ↗Diffusion strategies have advanced visual motor control by progressively denoising high-dimensional action sequences, providing a promising method for robot manipulation. However, as task complexity increases, the success rate of existing baseline models decreases considerably. Analysis indicates that current diffusion strategies are confronted with two limitations. First, these strategies only rely on short-term observations as conditions. Second, the training objective remains limited to a single denoising loss, which leads to error accumulation and causes grasping deviations. To address these limitations, this paper proposes Foresight-Conditioned Diffusion (ForeDiffusion), by injecting the predicted future view representation into the diffusion process. As a result, the policy is guided to be forward-looking, enabling it to correct trajectory deviations. Following this design, ForeDiffusion employs a dual loss mechanism, combining the traditional denoising loss and the consistency loss of future observations, to achieve the unified optimization. Extensive evaluation on the Adroit suite and the MetaWorld benchmark demonstrates that ForeDiffusion achieves an average success rate of 80% for the overall task, significantly outperforming the existing mainstream diffusion methods by approximately 20% in high difficulty tasks, while maintaining more stable performance across the entire tasks.
Vision-Language-Action (VLA) models typically bridge the gap between perceptual and action spaces by pre-training a large-scale Vision-Language Model (VLM) on robotic data. While this approach greatly enhances performance, it also incurs significant training costs. In this paper, we investigate how to effectively bridge vision-language (VL) representations to action (A). We introduce VLA-Adapter, a novel paradigm designed to reduce the reliance of VLA models on large-scale VLMs and extensive pre-training. To this end, we first systematically analyze the effectiveness of various VL conditions and present key findings on which conditions are essential for bridging perception and action spaces. Based on these insights, we propose a lightweight Policy module with Bridge Attention, which autonomously injects the optimal condition into the action space. In this way, our method achieves high performance using only a 0.5B-parameter backbone, without any robotic data pre-training. Extensive experiments on both simulated and real-world robotic benchmarks show that VLA-Adapter not only achieves state-of-the-art level performance, but also offers the fast inference speed reported to date. Furthermore, thanks to the proposed advanced bridging paradigm, VLA-Adapter enables the training of a powerful VLA model on a single consumer-grade GPU, greatly lowering the barrier to deploying VLA model.
Traditional language-conditioned manipulation agent adaptation to new manipulation skills leads to catastrophic forgetting of old skills, limiting dynamic scene practical deployment. In this paper, we propose SkillsCrafter, a novel robotic manipulation framework designed to continually learn multiple skills while reducing catastrophic forgetting of old skills. Specifically, we propose a Manipulation Skills Adaptation to retain the old skills knowledge while inheriting the shared knowledge between new and old skills to facilitate learning of new skills. Meanwhile, we perform the singular value decomposition on the diverse skill instructions to obtain common skill semantic subspace projection matrices, thereby recording the essential semantic space of skills. To achieve forget-less and generalization manipulation, we propose a Skills Specialization Aggregation to compute inter-skills similarity in skill semantic subspaces, achieving aggregation of the previously learned skill knowledge for any new or unknown skill. Extensive simulator and real-world experiments demonstrate the effectiveness and superiority of our SkillsCrafter.
Language-guided long-horizon mobile manipulation has long been a grand challenge in embodied semantic reasoning, generalizable manipulation, and adaptive locomotion. Three fundamental limitations hinder progress: First, although large language models have shown promise in enhancing spatial reasoning and task planning through learned semantic priors, existing implementations remain confined to tabletop scenarios, failing to address the constrained perception and limited actuation ranges characteristic of mobile platforms. Second, current manipulation strategies exhibit insufficient generalization when confronted with the diverse object configurations encountered in open-world environments. Third, while crucial for practical deployment, the dual requirement of maintaining high platform maneuverability alongside precise end-effector control in unstructured settings remains understudied in the literature. In this work, we present ODYSSEY, a unified mobile manipulation framework for agile quadruped robots equipped with manipulators, which seamlessly integrates high-level task planning with low-level whole-body control. To address the challenge of egocentric perception in language-conditioned tasks, we introduce a hierarchical planner powered by a vision-language model, enabling long-horizon instruction decomposition and precise action execution. At the control level, our novel whole-body policy achieves robust coordination of locomotion and manipulation across challenging terrains. We further present the first comprehensive benchmark for long-horizon mobile manipulation, evaluating diverse indoor and outdoor scenarios. Through successful sim-to-real transfer, we demonstrate the system’s generalization and robustness in real-world deployments, underscoring the practicality of legged manipulators in unstructured environments. Our work advances the feasibility of generalized robotic assistants capable of complex, dynamic tasks.
Current paradigms for robotic imitation learning face a stark trade-off between the motion fidelity of diffusion models and the data scalability of inverse dynamics models. The latter, while scalable, often learns a latent action space disconnected from physical reality. This flaw leads to critical failures: temporal entanglement, where the model cannot distinguish between visually similar states requiring distinct actions, e.g., a gripper approaching versus receding from an object. This ambiguity, compounded by discretization artifacts and sensitivity to task-irrelevant dynamics, renders robust planning infeasible. We introduce LatentVLA, a vision-language-action framework designed to overcome these limitations by learning a continuous and spatiotemporally grounded latent action representation. Its progressive three-stage architecture first employs a Temporal-Attentive Latent Action Model (TA-LAM) to resolve ambiguities using language-guided attention and explicit temporal encoding. Subsequently, a Latent Action Diffusion Transformer (LADT) performs planning via diffusion directly within this continuous latent space, preserving motion fidelity without tokenization. Finally, an expert policy head translates these latent plans into precise robot actions. Experiments show LatentVLA sets a new state-of-the-art across a suite of real-world bimanual tasks, outperforming prior methods and demonstrating superior zero-shot generalization and few-shot efficiency.
FARM: Frame-Accelerated Augmentation and Residual Mixture-of-Experts for Physics-Based High-Dynamic Humanoid Control
PDF ↗Unified physics-based humanoid controllers are pivotal for robotics and character animation, yet models that excel on gentle, everyday motions still stumble on explosive actions, hampering real-world deployment. We bridge this gap with FARM (Frame-Accelerated Augmentation and Residual Mixture-of-Experts), an end-to-end framework composed of frame-accelerated augmentation, a robust base controller, and a residual mixture-of-experts (MoE). Frame-accelerated augmentation exposes the model to high-velocity pose changes by widening inter-frame gaps. The base controller reliably tracks everyday low-dynamic motions, while the residual MoE adaptively allocates additional network capacity to handle challenging high-dynamic actions, significantly enhancing tracking accuracy. In the absence of a public benchmark, we curate the High-Dynamic Humanoid Motion (HDHM) dataset, comprising 3593 physically plausible clips. On HDHM, FARM reduces the tracking failure rate by 42.8% and lowers global mean per-joint position error by 14.6% relative to the baseline, while preserving near-perfect accuracy on low-dynamic motions. These results establish FARM as a new baseline for high-dynamic humanoid control and introduce the first open benchmark dedicated to this challenge.
ManipLVM-R1: Reinforcement Learning for Reasoning in Embodied Manipulation with Large Vision-Language Models
PDF ↗Large Vision-Language Models (LVLMs) have recently advanced robotic manipulation by leveraging vision for scene perception and language for instruction following. However, existing methods rely heavily on costly human-annotated training datasets, which limits their generalization and causes them to struggle in out-of-domain (OOD) scenarios, reducing real-world adaptability. To address these challenges, we propose ManipLVM-R1, a novel reinforcement learning framework that replaces traditional supervision with Reinforcement Learning using Verifiable Rewards (RLVR). By directly optimizing for task-aligned outcomes, our method enhances generalization and physical reasoning while removing the dependence on costly annotations. Specifically, we design two rule-based reward functions targeting key robotic manipulation subtasks: an Affordance Perception Reward to enhance localization of interaction regions, and a Trajectory Match Reward to ensure the physical plausibility of action paths. These rewards provide immediate feedback and impose spatial-logical constraints, encouraging the model to go beyond shallow pattern matching and instead learn deeper, more systematic reasoning about physical interactions. Experimental results show that ManipLVM-R1 achieves substantial performance gains across multiple manipulation tasks, using only 50% of the training data while achieving strong generalization to OOD scenarios. We further analyze the benefits of our reward design and its impact on task success and efficiency.
Recent advances in Vision-Language-Action (VLA) models have enabled robotic agents to integrate multimodal understanding with action execution. However, our empirical analysis reveals that current VLAs struggle to allocate visual attention to target regions. Instead, visual attention is always dispersed. To guide the visual attention grounding on the correct target, we propose ReconVLA, a reconstructive VLA model with an implicit grounding paradigm. Conditioned on the model's visual outputs, a diffusion transformer aims to reconstruct the gaze region of the image, which corresponds to the target manipulated objects. This process prompts the VLA model to learn fine-grained representations and accurately allocate visual attention, thus effectively leveraging task-specific visual information and conducting precise manipulation. Moreover, we curate a large-scale pretraining dataset comprising over 100k trajectories and 2 million data samples from open-source robotic datasets, further boosting the model’s generalization in visual reconstruction. Extensive experiments in simulation and the real world demonstrate the superiority of our implicit grounding method, showcasing its capabilities of precise manipulation and generalization.
Autonomous aerial robots must operate in cluttered, wind-disturbed environments where turbulence and gusts generated by wind-object and terrain interactions introduce significant aerodynamic risks, including orientation instability, sensor degradation, control drift, and increased power consumption, often leading to mission failure or crash. We present Graphlets-based Zero-Shot Planning Framework (GZS), a novel, non-parametric, fast computation, memory-efficient, zero-shot training-free onboard inference framework for real-time 3D spatial-aware aerodynamic risk perception that operates without prior scene knowledge. GZS dynamically classifies point clouds to extract local topology, incorporates physics-informed modeling of wind interactions, and applies attention-guided segment matching to generate onboard 3D representations of wind-induced aerodynamic risk. It transforms unstructured scene segments into structured graphlets topologies encoding aerodynamic risk-aware features, enabling UAVs to identify and navigate through regions of minimal aerodynamic hazard in real time and without prior training in any environment. Unlike computational fluid dynamics(CFD)-based, deep learning, or map-dependent approaches, GZS performs zero-shot aerodynamic risk estimation in previously unseen and dynamic conditions. Extensive experiments demonstrate 90-95% accurate aerodynamic risk zone identification compared to conventional methods of CFDs and wind tunnels, while substantially reducing computational and memory overhead, and a 100% success rate in creating onboard 3d spatial-aware risk perceptions. Our results establish GZS as a framework for a zero-shot, non-parametric, robust, aerodynamic risk perception for autonomous real-time trajectory planning in wind-affected aerial environments.
In robot manipulation, robot learning has become a prevailing approach. However, generative models within this field face a fundamental trade-off between the slow, iterative sampling of diffusion models and the architectural constraints of faster Flow-based methods, which often rely on explicit consistency losses. To address these limitations, we introduce MP1, which pairs 3D point-cloud inputs with the MeanFlow paradigm to generate action trajectories in one network function evaluation (1-NFE). By directly learning the interval-averaged velocity via the "MeanFlow Identity", our policy avoids any additional consistency constraints. This formulation eliminates numerical ODE-solver errors during inference, yielding more precise trajectories. MP1 further incorporates CFG for improved trajectory controllability while retaining 1-NFE inference without reintroducing structural constraints. Because subtle scene-context variations are critical for robot learning, especially in few-shot learning, we introduce a lightweight Dispersive Loss that repels state embeddings during training, boosting generalization without slowing inference. We validate our method on the Adroit and Meta-World benchmarks, as well as in real-world scenarios. Experimental results show MP1 achieves superior average task success rates, outperforming DP3 by 10.2% and FlowPolicy by 7.3%. Its average inference time is only 6.8 ms—19 times faster than DP3 and nearly 2 times faster than FlowPolicy.
The human nervous system exhibits bilateral symmetry, enabling coordinated and balanced movements. However, existing Deep Reinforcement Learning (DRL) methods for humanoid robots neglect morphological symmetry of the robot, leading to uncoordinated and suboptimal behaviors. Inspired by human motor control, we propose Symmetry Equivariant Policy (SE-Policy), a new DRL framework that embeds strict symmetry equivariance in the actor and symmetry invariance in the critic without additional hyperparameters. SE-Policy enforces consistent behaviors across symmetric observations, producing temporally and spatially coordinated motions with higher task performance. Extensive experiments on velocity tracking tasks, conducted in both simulation and real-world deployment with the Unitree G1 humanoid robot, demonstrate that SE-Policy improves tracking accuracy by up to 40% compared to state-of-the-art baselines, while achieving superior spatial-temporal coordination. These results demonstrate the effectiveness of SE-Policy and its broad applicability to humanoid robots.