Zhenyu Wei

Jinxin Liu, Ziqi Zhang, Zhenyu Wei et al.

Offline reinforcement learning (RL) aims to learn a policy using only pre-collected and fixed data. Although avoiding the time-consuming online interactions in RL, it poses challenges for out-of-distribution (OOD) state actions and often suffers from data inefficiency for training. Despite many efforts being devoted to addressing OOD state actions, the latter (data inefficiency) receives little attention in offline RL. To address this, this paper proposes the cross-domain offline RL, which assumes offline data incorporate additional source-domain data from varying transition dynamics (environments), and expects it to contribute to the offline data efficiency. To do so, we identify a new challenge of OOD transition dynamics, beyond the common OOD state actions issue, when utilizing cross-domain offline data. Then, we propose our method BOSA, which employs two support-constrained objectives to address the above OOD issues. Through extensive experiments in the cross-domain offline RL setting, we demonstrate BOSA can greatly improve offline data efficiency: using only 10\% of the target data, BOSA could achieve {74.4\%} of the SOTA offline RL performance that uses 100\% of the target data. Additionally, we also show BOSA can be effortlessly plugged into model-based offline RL and noising data augmentation techniques (used for generating source-domain data), which naturally avoids the potential dynamics mismatch between target-domain data and newly generated source-domain data.

Zhenyu Wei, Yunchao Yao, Mingyu Ding

Dexterous manipulation policies today largely assume fixed hand designs, severely restricting their generalization to new embodiments with varied kinematic and structural layouts. To overcome this limitation, we introduce a parameterized canonical representation that unifies a broad spectrum of dexterous hand architectures. It comprises a unified parameter space and a canonical URDF format, offering three key advantages. 1) The parameter space captures essential morphological and kinematic variations for effective conditioning in learning algorithms. 2) A structured latent manifold can be learned over our space, where interpolations between embodiments yield smooth and physically meaningful morphology transitions. 3) The canonical URDF standardizes the action space while preserving dynamic and functional properties of the original URDFs, enabling efficient and reliable cross-embodiment policy learning. We validate these advantages through extensive analysis and experiments, including grasp policy replay, VAE latent encoding, and cross-embodiment zero-shot transfer. Specifically, we train a VAE on the unified representation to obtain a compact, semantically rich latent embedding, and develop a grasping policy conditioned on the canonical representation that generalizes across dexterous hands. We demonstrate, through simulation and real-world tasks on unseen morphologies (e.g., 81.9% zero-shot success rate on 3-finger LEAP Hand), that our framework unifies both the representational and action spaces of structurally diverse hands, providing a scalable foundation for cross-hand learning toward universal dexterous manipulation. Project Page: https://zhenyuwei2003.github.io/OHRA/

Hongyin Zhang, Diyuan Shi, Zifeng Zhuang et al.

Developing robotic intelligent systems that can adapt quickly to unseen wild situations is one of the critical challenges in pursuing autonomous robotics. Although some impressive progress has been made in walking stability and skill learning in the field of legged robots, their ability for fast adaptation is still inferior to that of animals in nature. Animals are born with a massive set of skills needed to survive, and can quickly acquire new ones, by composing fundamental skills with limited experience. Inspired by this, we propose a novel framework, named Robot Skill Graph (RSG) for organizing a massive set of fundamental skills of robots and dexterously reusing them for fast adaptation. Bearing a structure similar to the Knowledge Graph (KG), RSG is composed of massive dynamic behavioral skills instead of static knowledge in KG and enables discovering implicit relations that exist in between the learning context and acquired skills of robots, serving as a starting point for understanding subtle patterns existing in robots' skill learning. Extensive experimental results demonstrate that RSG can provide reliable skill inference upon new tasks and environments, and enable quadruped robots to adapt to new scenarios and quickly learn new skills.

Yu Fang, Yuchun Feng, Dong Jing et al.

Vision-Language-Action models (VLAs) promise to ground language instructions in robot control, yet in practice often fail to faithfully follow language. When presented with instructions that lack strong scene-specific supervision, VLAs suffer from counterfactual failures: they act based on vision shortcuts induced by dataset biases, repeatedly executing well-learned behaviors and selecting objects frequently seen during training regardless of language intent. To systematically study it, we introduce LIBERO-CF, the first counterfactual benchmark for VLAs that evaluates language following capability by assigning alternative instructions under visually plausible LIBERO layouts. Our evaluation reveals that counterfactual failures are prevalent yet underexplored across state-of-the-art VLAs. We propose Counterfactual Action Guidance (CAG), a simple yet effective dual-branch inference scheme that explicitly regularizes language conditioning in VLAs. CAG combines a standard VLA policy with a language-unconditioned Vision-Action (VA) module, enabling counterfactual comparison during action selection. This design reduces reliance on visual shortcuts, improves robustness on under-observed tasks, and requires neither additional demonstrations nor modifications to existing architectures or pretrained models. Extensive experiments demonstrate its plug-and-play integration across diverse VLAs and consistent improvements. For example, on LIBERO-CF, CAG improves $π_{0.5}$ by 9.7% in language following accuracy and 3.6% in task success on under-observed tasks using a training-free strategy, with further gains of 15.5% and 8.5%, respectively, when paired with a VA model. In real-world evaluations, CAG reduces counterfactual failures of 9.4% and improves task success by 17.2% on average.

Dong Jing, Gang Wang, Jiaqi Liu et al.

Vision-language-action (VLA) models have shown remarkable capabilities in robotic manipulation, but their performance is sensitive to the $\textbf{action chunk length}$ used during training, termed $\textbf{horizon}$. Our empirical study reveals an inherent trade-off: longer horizons provide stronger global foresight but degrade fine-grained accuracy, while shorter ones sharpen local control yet struggle on long-term tasks, implying fixed choice of single horizons being suboptimal. To mitigate the trade-off, we propose a $\textbf{mixture of horizons (MoH)}$ strategy. MoH rearranges the action chunk into several segments with different horizons, processes them in parallel with a shared action transformer, and fuses outputs with a light linear gate. It has three appealing benefits. 1) MoH exploits long-term foresight and short-term precision jointly within a single model, improving both performance and generalizability to complex tasks. 2) MoH is plug-and-play for full-attention action modules with minimal training or inference overhead. 3) MoH enables dynamic inference with adaptive horizons, which selects stable actions through cross-horizon consensus, achieving 2.5$\times$ higher throughput than baselines while preserving superior performance. Extensive experiments over flow-based policies $π_0$, $π_{0.5}$, and one-step regression policy $π_{\text{reg}}$ demonstrate that MoH yields consistent and significant gains on both simulations and real-world tasks. Notably, under mixed-task setting, $π_{0.5}$ with MoH reaches a new state-of-the-art with 99$\%$ average success rate on LIBERO after only $30k$ training iterations. Project page: https://github.com/Timsty1/MixtureOfHorizons

Kun Lei, Huanyu Li, Dongjie Yu et al.

Real-world robotic manipulation in homes and factories demands reliability, efficiency, and robustness that approach or surpass the performance of skilled human operators. We present RL-100, a real-world reinforcement learning framework built on diffusion-based visuomotor policies. RL-100 unifies imitation and reinforcement learning under a single PPO-style objective applied within the denoising process, yielding conservative and stable policy improvements across both offline and online stages. To meet deployment latency constraints, we employ a lightweight consistency distillation procedure that compresses multi-step diffusion into a one-step controller for high-frequency control. The framework is task-, embodiment-, and representation-agnostic, and supports both single-action outputs and action-chunking control. We evaluate RL-100 on seven diverse real-robot manipulation tasks, ranging from dynamic pushing and agile bowling to pouring, cloth folding, unscrewing, and multi-stage juicing. RL-100 attains 100% success across evaluated trials, achieving 900 out of 900 successful episodes, including up to 250 out of 250 consecutive trials on one task, and matches or surpasses expert teleoperators in time-to-completion. Without retraining, a single policy attains approximately 90% zero-shot success under environmental and dynamics shifts, adapts in a few-shot regime to significant task variations (86.7%), and remains robust to aggressive human perturbations (about 95%). In a public shopping-mall deployment, the juicing robot served random customers continuously for roughly seven hours without failure. Together, these results suggest a practical path toward deployment-ready robot learning: start from human priors, align training objectives with human-grounded metrics, and reliably extend performance beyond human demonstrations.

Zhenyu Wei, Yujie He, Zhanchuan Cai

RGB-D tracking significantly improves the accuracy of object tracking. However, its dependency on real depth inputs and the complexity involved in multi-modal fusion limit its applicability across various scenarios. The utilization of depth information in RGB-D tracking inspired us to propose a new method, named MDETrack, which trains a tracking network with an additional capability to understand the depth of scenes, through supervised or self-supervised auxiliary Monocular Depth Estimation learning. The outputs of MDETrack's unified feature extractor are fed to the side-by-side tracking head and auxiliary depth estimation head, respectively. The auxiliary module will be discarded in inference, thus keeping the same inference speed. We evaluated our models with various training strategies on multiple datasets, and the results show an improved tracking accuracy even without real depth. Through these findings we highlight the potential of depth estimation in enhancing object tracking performance.

Feiyu Yao, Zongkai Wu, Zhenyu Wei et al.

Different application scenarios will cause IMU to exhibit different error characteristics which will cause trouble to robot application. However, most data processing methods need to be designed for specific scenario. To solve this problem, we propose a few-shot domain adaptation method. In this work, a domain adaptation framework is considered for denoising the IMU, a reconstitution loss is designed to improve domain adaptability. In addition, in order to further improve the adaptability in the case of limited data, a few-shot training strategy is adopted. In the experiment, we quantify our method on two datasets (EuRoC and TUM-VI) and two real robots (car and quadruped robot) with three different precision IMUs. According to the experimental results, the adaptability of our framework is verified by t-SNE. In orientation results, our proposed method shows the great denoising performance.