arXiv subjects

Yuheng Zhou

Publications and source records attributed to Yuheng Zhou.

10 recordsLinked to original sources

arXiv2026

WA-JEPA: Rethinking the Video JEPA Paradigm for World-Action Modeling in Autonomous Driving

Video Joint Embedding Predictive Architecture (V-JEPA) learns powerful spatiotemporal representations from video through self-supervised latent feature prediction. However, V-JEPA is built around random-mask completion and deterministic regression, making it fundamentally ill-suited for autonomous driving planning that demands future-directed prediction tightly coupled with action. To address this, we rethink the V-JEPA paradigm and present WA-JEPA, a V-JEPA-native world-action model designed for autonomous driving planning. Instead of random spatiotemporal masking, WA-JEPA employs hybrid future-masked pre-training, where the model infers future latents from observed context. Departing from deterministic regression, we recast future prediction as conditional flow matching over latent futures, which substantially improves the model's ability to generate plausible future latents for downstream planning. Finally, a joint future-action predictor is proposed to denoise future scene tokens and ego trajectories together in a unified spatiotemporal latent space, allowing action supervision to directly shape planning-relevant world representations. Pre-trained on nuPlan videos and fine-tuned on NAVSIM, WA-JEPA reaches 91.7 EPDMS on NAVSIM-v2, surpassing the strongest end-to-end and world-action baselines by 1.6 and 1.3 EPDMS, and, without HUGSIM-specific fine-tuning, attains the best HD-Score of 0.4462 on the closed-loop HUGSIM benchmark under the same evaluation protocol. These results validate V-JEPA-native world-action modeling as a powerful and scalable paradigm for autonomous driving planning. Code is available at https://github.com/AFARI-Research/WA-JEPA.

cs.CV↗

arXiv2026

Conditional grain-graph diffusion for property-guided inverse design of polycrystalline microstructures

Graph representations compactly encode polycrystalline microstructures while retaining grain topology and grain boundary information. We present a conditional graph diffusion framework for property-guided inverse design of dual-phase Ti-6Al-4V microstructures. An enhanced grain graph neural network (GNN) with grain boundary edge features, learnable node and edge embeddings, and multi-statistic pooling serves as a forward surrogate for stress prediction and candidate evaluation. The conditional diffusion model generates candidates through reverse diffusion under prescribed {\alpha}-phase volume fraction, elastic modulus, and yield-stress proxy targets. Across four target regimes and independently seeded starting sets, generated candidates consistently approach the prescribed properties, including a target outside the property envelope of the existing microstructures. Local crystallographic consistency is evaluated post-generation from deviations from the Burgers orientation relationship (BOR). BOR-aware ranking increases mean BOR consistency by up to 44.9% and 56.4% for the in- and out-of-envelope targets, respectively, while maintaining property alignment. Finite element validation of the five best candidates in each primary design case yields a maximum absolute relative error of 1.0% in their mean properties. In a representative benchmark, diffusion requires 32 candidate evaluations per input graph, compared with approximately 40,000 for random search and evolutionary optimization, and reduces runtime by approximately two orders of magnitude in the tested implementations. These results establish conditional grain-graph diffusion as an efficient framework for property-guided polycrystalline microstructure design.

cond-mat.mtrl-sci↗

arXiv2026

CoWorld-VLA: Thinking in a Multi-Expert World Model for Autonomous Driving

Vision-Language-Action (VLA) models have emerged as a promising paradigm for end-to-end autonomous driving. However, existing reasoning mechanisms still struggle to provide planning-oriented intermediate representations: textual Chain-of-Thought (CoT) fails to preserve continuous spatiotemporal structure, while latent world reasoning remains difficult to use as a direct condition for action generation. In this paper, we propose CoWorld-VLA, a multi-expert world reasoning framework for autonomous driving, where world representations serve as explicit conditions to guide action planning. CoWorld-VLA extracts complementary world information through multi-source supervision and encodes it into expert tokens within the VLA, thereby providing planner-accessible conditioning signals. Specifically, we construct four types of tokens: semantic interaction, geometric structure, dynamic evolution, and ego trajectory tokens, which respectively model interaction intent, spatial structure, future temporal dynamics, and behavioral goals. During action generation, CoWorld-VLA employs a diffusion-based hierarchical multi-expert fusion planner, which is coupled with scene context throughout the joint denoising process to generate continuous ego trajectories. Experiments on NAVSIM v1 & v2 demonstrate future-scene modeling capability and strong planning performance, including collision avoidance and trajectory accuracy. Ablation studies further validate the complementarity of expert tokens and their effectiveness as planning conditions for action generation. Code will be available at https://github.com/AFARI-Research/CoWorld-VLA.

cs.CV↗

arXiv2026

DockAnywhere: Data-Efficient Visuomotor Policy Learning for Mobile Manipulation via Novel Demonstration Generation

Mobile manipulation is a fundamental capability that enables robots to interact in expansive environments such as homes and factories. Most existing approaches follow a two-stage paradigm, where the robot first navigates to a docking point and then performs fixed-base manipulation using powerful visuomotor policies. However, real-world mobile manipulation often suffers from the view generalization problem due to shifts of docking points. To address this issue, we propose a novel low-cost demonstration generation framework named DockAnywhere, which improves viewpoint generalization under docking variability by lifting a single demonstration to diverse feasible docking configurations. Specifically, DockAnywhere lifts a trajectory to any feasible docking points by decoupling docking-dependent base motions from contact-rich manipulation skills that remain invariant across viewpoints. Feasible docking proposals are sampled under feasibility constraints, and corresponding trajectories are generated via structure-preserving augmentation. Visual observations are synthesized in 3D space by representing the robot and objects as point clouds and applying point-level spatial editing to ensure the consistency of observation and action across viewpoints. Extensive experiments on ManiSkill and real-world platforms demonstrate that DockAnywhere substantially improves policy success rates and easily generalizes to novel viewpoints from unseen docking points during training, significantly enhancing the generalization capability of mobile manipulation policy in real-world deployment.

cs.RO↗

arXiv2026

RoCo Challenge at AAAI 2026: Benchmarking Robotic Collaborative Manipulation for Assembly Towards Industrial Automation

Embodied Artificial Intelligence (EAI) is rapidly developing, gradually subverting previous autonomous systems' paradigms from isolated perception to integrated, continuous action. This transition is highly significant for industrial robotic manipulation, promising to free human workers from repetitive, dangerous daily labor. To benchmark and advance this capability, we introduce the Robotic Collaborative Assembly Assistance (RoCo) Challenge with a dataset towards simulation and real-world assembly manipulation. Set against the backdrop of human-centered manufacturing, this challenge focuses on a high-precision planetary gearbox assembly task, a demanding yet highly representative operation in modern industry. Built upon a self-developed data collection, training, and evaluation system in Isaac Sim, and utilizing a dual-arm robot for real-world deployment, the challenge operates in two phases. The Simulation Round defines fine-grained task phases for step-wise scoring to handle the long-horizon nature of the assembly. The Real-World Round mirrors this evaluation with physical gearbox components and high-quality teleoperated datasets. The core tasks require assembling an epicyclic gearbox from scratch, including mounting three planet gears, a sun gear, and a ring gear. Attracting over 60 teams and 170+ participants from more than 10 countries, the challenge yielded highly effective solutions, most notably ARC-VLA and RoboCola. Results demonstrate that a dual-model framework for long-horizon multi-task learning is highly effective, and the strategic utilization of recovery-from-failure curriculum data is a critical insight for successful deployment. This report outlines the competition setup, evaluation approach, key findings, and future directions for industrial EAI. Our dataset, CAD files, code, and evaluation results can be found at: https://rocochallenge.github.io/RoCo2026/.

cs.RO↗

arXiv2026

UniManip: General-Purpose Zero-Shot Robotic Manipulation with Agentic Operational Graph

Achieving general-purpose robotic manipulation requires robots to seamlessly bridge high-level semantic intent with low-level physical interaction in unstructured environments. However, existing approaches falter in zero-shot generalization: end-to-end Vision-Language-Action (VLA) models often lack the precision required for long-horizon tasks, while traditional hierarchical planners suffer from semantic rigidity when facing open-world variations. To address this, we present UniManip, a framework grounded in a Bi-level Agentic Operational Graph (AOG) that unifies semantic reasoning and physical grounding. By coupling a high-level Agentic Layer for task orchestration with a low-level Scene Layer for dynamic state representation, the system continuously aligns abstract planning with geometric constraints, enabling robust zero-shot execution. Unlike static pipelines, UniManip operates as a dynamic agentic loop: it actively instantiates object-centric scene graphs from unstructured perception, parameterizes these representations into collision-free trajectories via a safety-aware local planner, and exploits structured memory to autonomously diagnose and recover from execution failures. Extensive experiments validate the system's robust zero-shot capability on unseen objects and tasks, demonstrating a 22.5% and 25.0% higher success rate compared to state-of-the-art VLA and hierarchical baselines, respectively. Notably, the system enables direct zero-shot transfer from fixed-base setups to mobile manipulation without fine-tuning or reconfiguration. Our open-source project page can be found at https://henryhcliu.github.io/unimanip.

cs.RO↗

arXiv2025

VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning

Recent high-capacity vision-language-action (VLA) models have demonstrated impressive performance on a range of robotic manipulation tasks by imitating human demonstrations. However, exploiting offline data with limited visited states will cause execution failure in out-of-distribution scenarios. Intuitively, an exploration-based method that improves on online collected data at test time could address this limitation. We present VLA-RL, an algorithmic and systematic framework that leverages online reinforcement learning (RL) to improve pretrained auto-regressive VLAs in downstream tasks. Within a unified perspective, we first introduce a trajectory-level RL formulation for auto-regressive VLA training, which models general robotic manipulation trajectory as multi-modal multi-turn conversation. To address the challenge of sparse rewards, we fine-tune a pretrained vision-language model as a robotic process reward model, which is trained on pseudo reward labels annotated on automatically extracted task segments. To scale up, we identify several implementation findings that improve the stability and efficiency including curriculum selection strategy, GPU-balanced vectorized environments, batch decoding, and critic warmup. VLA-RL enables OpenVLA-7B to surpass the strongest finetuned baseline by 4.5% on 40 challenging robotic manipulation tasks in LIBERO, and even matches the performance of advanced commercial models such as $\pi_0$-FAST. Notably, we observe that VLA-RL benefits from increased test-time optimization, indicating an early spark of inference scaling laws in robotics.

cs.RO↗

arXiv2025

A spectral Levenberg-Marquardt-Deflation method for multiple solutions of semilinear elliptic systems

Many nonlinear differential equations arising from practical problems may permit nontrivial multiple solutions relevant to applications, and these multiple solutions are helpful to deeply understand these practical problems and to improve some applications. Developing an efficient numerical method for finding multiple solutions is very necessary due to the nonlinearity and multiple solutions of these equations. Moreover, providing an efficient iteration plays an important role in successfully obtaining multiple solutions with fast and stable convergence. In the current paper, an efficient algorithm for finding multiple solutions of semilinear elliptic systems is proposed, where the trust region Levenberg-Marquardt method is firstly used to iterate the resulted nonlinear algebraic system. When the nonlinear term in these equations has only the first derivative, our algorithm can efficiently find multiple solutions as well. Several numerical experiments are tested to show the efficiency of our algorithm, and some solutions which have not been shown in the literature are also found and shown.

math.OC↗

arXiv2025

MoManipVLA: Transferring Vision-language-action Models for General Mobile Manipulation

Mobile manipulation is the fundamental challenge for robotics to assist humans with diverse tasks and environments in everyday life. However, conventional mobile manipulation approaches often struggle to generalize across different tasks and environments because of the lack of large-scale training. In contrast, recent advances in vision-language-action (VLA) models have shown impressive generalization capabilities, but these foundation models are developed for fixed-base manipulation tasks. Therefore, we propose an efficient policy adaptation framework named MoManipVLA to transfer pre-trained VLA models of fix-base manipulation to mobile manipulation, so that high generalization ability across tasks and environments can be achieved in mobile manipulation policy. Specifically, we utilize pre-trained VLA models to generate waypoints of the end-effector with high generalization ability. We design motion planning objectives for the mobile base and the robot arm, which aim at maximizing the physical feasibility of the trajectory. Finally, we present an efficient bi-level objective optimization framework for trajectory generation, where the upper-level optimization predicts waypoints for base movement to enhance the manipulator policy space, and the lower-level optimization selects the optimal end-effector trajectory to complete the manipulation task. In this way, MoManipVLA can adjust the position of the robot base in a zero-shot manner, thus making the waypoints predicted from the fixed-base VLA models feasible. Extensive experimental results on OVMM and the real world demonstrate that MoManipVLA achieves a 4.2% higher success rate than the state-of-the-art mobile manipulation, and only requires 50 training cost for real world deployment due to the strong generalization ability in the pre-trained VLA models.

cs.RO↗

arXiv2025

Don't Confuse! Redrawing GUI Navigation Flow in Mobile Apps for Visually Impaired Users

Mobile applications (apps) are integral to our daily lives, offering diverse services and functionalities. They enable sighted users to access information coherently in an extremely convenient manner. However, it remains unclear if visually impaired users, who rely solely on the screen readers (e.g., Talkback) to navigate and access app information, can do so in the correct and reasonable order. This may result in significant information bias and operational errors. Considering these issues, in this work, we proposed a method named RGNF (Re-draw GUI Navigation Flow). It aimed to enhance the understandability and coherence of accessing the content of each component within the Graphical User Interface (GUI), together with assisting developers in creating well-designed GUI navigation flow (GNF). This method was inspired by the characteristics identified in our preliminary study, where visually impaired users expected navigation to be associated with close position and similar shape of GUI components that were read consecutively. Thus, our method relied on the principles derived from the Gestalt psychological model, aiming to group GUI components into different regions according to the laws of proximity and similarity, thereby redrawing the GNFs. To evaluate the effectiveness of our method, we calculated sequence similarity values before and after redrawing the GNF, and further employed the tools proposed by Alotaibi et al. to measure the reachability of GUI components. Our results demonstrated a substantial improvement in similarity (0.921) compared to the baseline (0.624), together with the reachability (90.31%) compared to the baseline GNF (74.35%). Furthermore, a qualitative user study revealed that our method had a positive effect on providing visually impaired users with an improved user experience.

cs.SE↗