Ce Hao

Kailin Lyu, Zhiqiang Yuan, Jianwei He et al.

Deep Image Search requires multi-step reasoning over rich contextual cues, such as time, location, and event relations. However, most existing LLM-based agents are stateless and reactive, lacking persistent memory to maintain long-horizon context or transfer experience across tasks, which often leads to execution drift and experience isolation. To address these limitations, we propose PhotoCraft, a training-free, hierarchical memory system for photo-search agents. Inspired by human cognition, PhotoCraft equips MLLMs with working, episodic, and semantic memory, which are dynamically invoked during reasoning to preserve logical consistency and knowledge transferability throughout multi-step reasoning and answer generation. Extensive experiments on DISBench demonstrate that PhotoCraft consistently improves context-aware retrieval across diverse MLLM backbones, achieving gains of up to 18.5\% and effectively mitigating key bottlenecks in memoryless deep image search, offering a practical path toward reliable and generalizable multimodal search agents.

Ce Hao, Catherine Weaver, Chen Tang et al.

Hierarchical reinforcement learning (RL) can accelerate long-horizon decision-making by temporally abstracting a policy into multiple levels. Promising results in sparse reward environments have been seen with skills, i.e. sequences of primitive actions. Typically, a skill latent space and policy are discovered from offline data. However, the resulting low-level policy can be unreliable due to low-coverage demonstrations or distribution shifts. As a solution, we propose the Skill-Critic algorithm to fine-tune the low-level policy in conjunction with high-level skill selection. Our Skill-Critic algorithm optimizes both the low-level and high-level policies; these policies are initialized and regularized by the latent space learned from offline demonstrations to guide the parallel policy optimization. We validate Skill-Critic in multiple sparse-reward RL environments, including a new sparse-reward autonomous racing task in Gran Turismo Sport. The experiments show that Skill-Critic's low-level policy fine-tuning and demonstration-guided regularization are essential for good performance. Code and videos are available at our website: https://sites.google.com/view/skill-critic.

Kailin Lyu, Kangyi Wu, Pengna Li et al.

LLM-based agents have demonstrated impressive zero-shot performance in vision-language navigation (VLN) tasks. However, most zero-shot methods primarily rely on closed-source LLMs as navigators, which face challenges related to high token costs and potential data leakage risks. Recent efforts have attempted to address this by using open-source LLMs combined with a spatiotemporal CoT framework, but they still fall far short compared to closed-source models. In this work, we identify a critical issue, Navigation Amnesia, through a detailed analysis of the navigation process. This issue leads to navigation failures and amplifies the gap between open-source and closed-source methods. To address this, we propose HiMemVLN, which incorporates a Hierarchical Memory System into a multimodal large model to enhance visual perception recall and long-term localization, mitigating the amnesia issue and improving the agent's navigation performance. Extensive experiments in both simulated and real-world environments demonstrate that HiMemVLN achieves nearly twice the performance of the open-source state-of-the-art method. The code is available at https://github.com/lvkailin0118/HiMemVLN.

Qiaojun Yu, Junbo Wang, Wenhai Liu et al.

Articulated objects like cabinets and doors are widespread in daily life. However, directly manipulating 3D articulated objects is challenging because they have diverse geometrical shapes, semantic categories, and kinetic constraints. Prior works mostly focused on recognizing and manipulating articulated objects with specific joint types. They can either estimate the joint parameters or distinguish suitable grasp poses to facilitate trajectory planning. Although these approaches have succeeded in certain types of articulated objects, they lack generalizability to unseen objects, which significantly impedes their application in broader scenarios. In this paper, we propose a novel framework of Generalizable Articulation Modeling and Manipulating for Articulated Objects (GAMMA), which learns both articulation modeling and grasp pose affordance from diverse articulated objects with different categories. In addition, GAMMA adopts adaptive manipulation to iteratively reduce the modeling errors and enhance manipulation performance. We train GAMMA with the PartNet-Mobility dataset and evaluate with comprehensive experiments in SAPIEN simulation and real-world Franka robot. Results show that GAMMA significantly outperforms SOTA articulation modeling and manipulation algorithms in unseen and cross-category articulated objects. We will open-source all codes and datasets in both simulation and real robots for reproduction in the final version. Images and videos are published on the project website at: http://sites.google.com/view/gamma-articulation

Linbo Wang, Yupeng Zheng, Qiang Chen et al. · tsinghua

We introduce Latent-WAM, an efficient end-to-end autonomous driving framework that achieves strong trajectory planning through spatially-aware and dynamics-informed latent world representations. Existing world-model-based planners suffer from inadequately compressed representations, limited spatial understanding, and underutilized temporal dynamics, resulting in sub-optimal planning under constrained data and compute budgets. Latent-WAM addresses these limitations with two core modules: a Spatial-Aware Compressive World Encoder (SCWE) that distills geometric knowledge from a foundation model and compresses multi-view images into compact scene tokens via learnable queries, and a Dynamic Latent World Model (DLWM) that employs a causal Transformer to autoregressively predict future world status conditioned on historical visual and motion representations. Extensive experiments on NAVSIM v2 and HUGSIM demonstrate new state-of-the-art results: 89.3 EPDMS on NAVSIM v2 and 28.9 HD-Score on HUGSIM, surpassing the best prior perception-free method by 3.2 EPDMS with significantly less training data and a compact 104M-parameter model.

Xin Li, Siyuan Huang, Qiaojun Yu et al.

Automating garment manipulation poses a significant challenge for assistive robotics due to the diverse and deformable nature of garments. Traditional approaches typically require separate models for each garment type, which limits scalability and adaptability. In contrast, this paper presents a unified approach using vision-language models (VLMs) to improve keypoint prediction across various garment categories. By interpreting both visual and semantic information, our model enables robots to manage different garment states with a single model. We created a large-scale synthetic dataset using advanced simulation techniques, allowing scalable training without extensive real-world data. Experimental results indicate that the VLM-based method significantly enhances keypoint detection accuracy and task success rates, providing a more flexible and general solution for robotic garment manipulation. In addition, this research also underscores the potential of VLMs to unify various garment manipulation tasks within a single framework, paving the way for broader applications in home automation and assistive robotics for future.

Yuhang Zheng, Songen Gu, Weize Li et al.

Contact-rich manipulation tasks, such as wiping and assembly, require accurate perception of contact forces, friction changes, and state transitions that cannot be reliably inferred from vision alone. Despite growing interest in visuo-tactile manipulation, progress is constrained by two persistent limitations: existing datasets are small in scale and narrow in task coverage, and current methods treat tactile signals as passive observations rather than using them to model contact dynamics or enable closed-loop control explicitly. In this paper, we present \textbf{OmniViTac}, a large-scale visuo-tactile-action dataset comprising $21{,}000+$ trajectories across $86$ tasks and $100+$ objects, organized into six physics-grounded interaction patterns. Building on this dataset, we propose \textbf{OmniVTA}, a world-model-based visuo-tactile manipulation framework that integrates four tightly coupled modules: a self-supervised tactile encoder, a two-stream visuo-tactile world model for predicting short-horizon contact evolution, a contact-aware fusion policy for action generation, and a 60Hz reflexive controller that corrects deviations between predicted and observed tactile signals in a closed loop. Real-robot experiments across all six interaction categories show that OmniVTA outperforms existing methods and generalizes well to unseen objects and geometric configurations, confirming the value of combining predictive contact modeling with high-frequency tactile feedback for contact-rich manipulation. All data, models, and code will be made publicly available on the project website at https://mrsecant.github.io/OmniVTA.

Qiaojun Yu, Ce Hao, Junbo Wang et al.

Robotic manipulation in everyday scenarios, especially in unstructured environments, requires skills in pose-aware object manipulation (POM), which adapts robots' grasping and handling according to an object's 6D pose. Recognizing an object's position and orientation is crucial for effective manipulation. For example, if a mug is lying on its side, it's more effective to grasp it by the rim rather than the handle. Despite its importance, research in POM skills remains limited, because learning manipulation skills requires pose-varying simulation environments and datasets. This paper introduces ManiPose, a pioneering benchmark designed to advance the study of pose-varying manipulation tasks. ManiPose encompasses: 1) Simulation environments for POM feature tasks ranging from 6D pose-specific pick-and-place of single objects to cluttered scenes, further including interactions with articulated objects. 2) A comprehensive dataset featuring geometrically consistent and manipulation-oriented 6D pose labels for 2936 real-world scanned rigid objects and 100 articulated objects across 59 categories. 3) A baseline for POM, leveraging the inferencing abilities of LLM (e.g., ChatGPT) to analyze the relationship between 6D pose and task-specific requirements, offers enhanced pose-aware grasp prediction and motion planning capabilities. Our benchmark demonstrates notable advancements in pose estimation, pose-aware manipulation, and real-robot skill transfer, setting new standards for POM research. We will open-source the ManiPose benchmark with the final version paper, inviting the community to engage with our resources, available at our website:https://sites.google.com/view/manipose.

Jianxin Bi, Kevin Yuchen Ma, Ce Hao et al.

Tactile feedback is generally recognized to be crucial for effective interaction with the physical world. However, state-of-the-art Vision-Language-Action (VLA) models lack the ability to interpret and use tactile signals, limiting their effectiveness in contact-rich tasks. Incorporating tactile feedback into these systems is challenging due to the absence of large multi-modal datasets. We present VLA-Touch, an approach that enhances generalist robot policies with tactile sensing \emph{without fine-tuning} the base VLA. Our method introduces two key innovations: (1) a pipeline that leverages a pretrained tactile-language model that provides semantic tactile feedback for high-level task planning, and (2) a diffusion-based controller that refines VLA-generated actions with tactile signals for contact-rich manipulation. Through real-world experiments, we demonstrate that our dual-level integration of tactile feedback improves task planning efficiency while enhancing execution precision. Code is open-sourced at \href{https://github.com/jxbi1010/VLA-Touch}{this URL}.

Yang Li, Zhaxizhuoma, Hongru Jiang et al.

Embodied intelligence for contact-rich manipulation has predominantly relied on position control, while explicit awareness and regulation of interaction forces remain under-explored, limiting stability, precision, and robustness in real-world tasks. We propose ForceVLA2, an end-to-end vision-language-action framework that equips robots with hybrid force-position control and explicit force awareness. ForceVLA2 introduces force-based prompts into the VLM expert to construct force-aware task concepts across stages, and employs a Cross-Scale Mixture-of-Experts (MoE) in the action expert to adaptively fuse these concepts with real-time interaction forces for closed-loop hybrid force-position regulation. To support learning and evaluation, we construct ForceVLA2-Dataset, containing 1,000 trajectories over 5 contact-rich tasks, including wiping, pressing, and assembling, with multi-view images, task prompts, proprioceptive state, and force signals. Extensive experiments show that ForceVLA2 substantially improves success rates and reliability in contact-rich manipulation, outperforming pi0 and pi0.5 by 48.0% and 35.0%, respectively, across the 5 tasks, and mitigating common failure modes such as arm overload and unstable contact, thereby actively advancing force-aware interactive physical intelligence in VLAs. The project page is available at https://sites.google.com/view/force-vla2/home.

Catherine Weaver, Chen Tang, Ce Hao et al.

Imitation learning learns a policy from demonstrations without requiring hand-designed reward functions. In many robotic tasks, such as autonomous racing, imitated policies must model complex environment dynamics and human decision-making. Sequence modeling is highly effective in capturing intricate patterns of motion sequences but struggles to adapt to new environments or distribution shifts that are common in real-world robotics tasks. In contrast, Adversarial Imitation Learning (AIL) can mitigate this effect, but struggles with sample inefficiency and handling complex motion patterns. Thus, we propose BeTAIL: Behavior Transformer Adversarial Imitation Learning, which combines a Behavior Transformer (BeT) policy from human demonstrations with online AIL. BeTAIL adds an AIL residual policy to the BeT policy to model the sequential decision-making process of human experts and correct for out-of-distribution states or shifts in environment dynamics. We test BeTAIL on three challenges with expert-level demonstrations of real human gameplay in Gran Turismo Sport. Our proposed residual BeTAIL reduces environment interactions and improves racing performance and stability, even when the BeT is pretrained on different tracks than downstream learning. Videos and code available at: https://sites.google.com/berkeley.edu/BeTAIL/home.

Puming Jiang, Meiyi Wang, Kelvin Lin et al.

Inference-time guidance is essential for steering generative robot policies toward dynamic objectives without retraining, yet existing methods are largely confined to chunk-based architectures that exhibit high latency and lack the reactivity needed for test-time preference alignment or obstacle avoidance. In this work, we formally derive the optimal guidance term for Stochastic Interpolants (SI) by analyzing the value function's time evolution via the Backward Kolmogorov Equation, establishing a modified drift that theoretically guarantees sampling from a target distribution. We apply this framework to real-time control through the Streaming Stochastic Interpolant Policy (SSIP), which generalizes the deterministic Streaming Flow Policy (SFP). Unifying this guidance law with the streaming architecture enables fast and reactive control. To support diverse deployment needs, we propose two complementary mechanisms: training-free Stochastic Trajectory Ensemble Guidance (STEG) that computes gradients on-the-fly for zero-shot adaptation, and training-based Conditional Critic Guidance (CCG) for amortized inference. Empirical evaluations demonstrate that our guided streaming approach significantly outperforms conventional chunk-based policies in reactivity and provides superior, physically valid guidance for dynamic, unstructured environments.

Jiawen Yu, Hairuo Liu, Qiaojun Yu et al.

Vision-Language-Action (VLA) models have advanced general-purpose robotic manipulation by leveraging pretrained visual and linguistic representations. However, they struggle with contact-rich tasks that require fine-grained control involving force, especially under visual occlusion or dynamic uncertainty. To address these limitations, we propose ForceVLA, a novel end-to-end manipulation framework that treats external force sensing as a first-class modality within VLA systems. ForceVLA introduces FVLMoE, a force-aware Mixture-of-Experts fusion module that dynamically integrates pretrained visual-language embeddings with real-time 6-axis force feedback during action decoding. This enables context-aware routing across modality-specific experts, enhancing the robot's ability to adapt to subtle contact dynamics. We also introduce \textbf{ForceVLA-Data}, a new dataset comprising synchronized vision, proprioception, and force-torque signals across five contact-rich manipulation tasks. ForceVLA improves average task success by 23.2% over strong pi_0-based baselines, achieving up to 80% success in tasks such as plug insertion. Our approach highlights the importance of multimodal integration for dexterous manipulation and sets a new benchmark for physically intelligent robotic control. Code and data will be released at https://sites.google.com/view/forcevla2025.

Yongen Zhao, Zihao Xu, Wenzhi Lu et al.

Wheeled-legged robots combine the energy efficiency of wheeled locomotion with the terrain adaptability of legged systems, making them promising platforms for agile mobility in complex and dynamic environments. However, enabling high-dynamic reflexive evasion against fast-moving obstacles remains challenging due to the hybrid morphology, mode coupling, and non-holonomic constraints of such platforms. In this work, we propose AWARE, Adaptive Wheeled-Legged Avoidance and Reflexive Evasion, a hierarchical reinforcement learning framework for high-dynamic obstacle avoidance in wheeled-legged robots. The proposed system naturally exhibits diverse emergent gaits and evasive behaviors, including forward lunge and lateral dodge, thereby leveraging the robot's hybrid morphology to enhance agility under highly dynamic threats. Extensive experiments in Isaac Lab simulation and real-world deployment on the M20 platform across diverse dynamic scenarios demonstrate that AWARE achieves robust and agile obstacle avoidance while revealing behaviorally distinct evasive strategies. These results highlight both the practical effectiveness of AWARE and the intrinsic reflexive agility of wheeled-legged robots.

Ce Hao, Anxing Xiao, Zhiwei Xue et al.

Diffusion-based planners have shown strong performance in short-horizon tasks but often fail in complex, long-horizon settings. We trace the failure to loose coupling between high-level (HL) sub-goal selection and low-level (LL) trajectory generation, which leads to incoherent plans and degraded performance. We propose Coupled Hierarchical Diffusion (CHD), a framework that models HL sub-goals and LL trajectories jointly within a unified diffusion process. A shared classifier passes LL feedback upstream so that sub-goals self-correct while sampling proceeds. This tight HL-LL coupling improves trajectory coherence and enables scalable long-horizon diffusion planning. Experiments across maze navigation, tabletop manipulation, and household environments show that CHD consistently outperforms both flat and hierarchical diffusion baselines. Our website is: https://sites.google.com/view/chd2025/home

Qiaojun Yu, Ce Hao, Xibin Yuan et al.

Manipulating articulated objects with robotic arms is challenging due to the complex kinematic structure, which requires precise part segmentation for efficient manipulation. In this work, we introduce a novel superpoint-based perception method designed to improve part segmentation in 3D point clouds of articulated objects. We propose a learnable, part-aware superpoint generation technique that efficiently groups points based on their geometric and semantic similarities, resulting in clearer part boundaries. Furthermore, by leveraging the segmentation capabilities of the 2D foundation model SAM, we identify the centers of pixel regions and select corresponding superpoints as candidate query points. Integrating a query-based transformer decoder further enhances our method's ability to achieve precise part segmentation. Experimental results on the GAPartNet dataset show that our method outperforms existing state-of-the-art approaches in cross-category part segmentation, achieving AP50 scores of 77.9% for seen categories (4.4% improvement) and $39.3\%$ for unseen categories (11.6% improvement), with superior results in 5 out of 9 part categories for seen objects and outperforming all previous methods across all part categories for unseen objects.

Xuanran Zhai, Qianyou Zhao, Qiaojun Yu et al.

Flow-matching-based policies have recently emerged as a promising approach for learning-based robot manipulation, offering significant acceleration in action sampling compared to diffusion-based policies. However, conventional flow-matching methods struggle with multi-modality, often collapsing to averaged or ambiguous behaviors in complex manipulation tasks. To address this, we propose the Variational Flow-Matching Policy (VFP), which introduces a variational latent prior for mode-aware action generation and effectively captures both task-level and trajectory-level multi-modality. VFP further incorporates Kantorovich Optimal Transport (K-OT) for distribution-level alignment and utilizes a Mixture-of-Experts (MoE) decoder for mode specialization and efficient inference. We comprehensively evaluate VFP on 41 simulated tasks and 3 real-robot tasks, demonstrating its effectiveness and sampling efficiency in both simulated and real-world settings. Results show that VFP achieves a 49% relative improvement in task success rate over standard flow-based baselines in simulation, and further outperforms them on real-robot tasks, while still maintaining fast inference and a compact model size. More details are available on our project page: https://sites.google.com/view/varfp/

Zhenmin Huang, Ce Hao, Wei Zhan et al.

Autonomous racing has gained significant attention as a platform for high-speed decision-making and motion control. While existing methods primarily focus on trajectory planning and overtaking strategies, the role of sportsmanship in ensuring fair competition remains largely unexplored. In human racing, rules such as the one-motion rule and the enough-space rule prevent dangerous and unsportsmanlike behavior. However, autonomous racing systems often lack mechanisms to enforce these principles, potentially leading to unsafe maneuvers. This paper introduces a bi-level game-theoretic framework to integrate sportsmanship (SPS) into versus racing. At the high level, we model racing intentions using a Stackelberg game, where Monte Carlo Tree Search (MCTS) is employed to derive optimal strategies. At the low level, vehicle interactions are formulated as a Generalized Nash Equilibrium Problem (GNEP), ensuring that all agents follow sportsmanship constraints while optimizing their trajectories. Simulation results demonstrate the effectiveness of the proposed approach in enforcing sportsmanship rules while maintaining competitive performance. We analyze different scenarios where attackers and defenders adhere to or disregard sportsmanship rules and show how knowledge of these constraints influences strategic decision-making. This work highlights the importance of balancing competition and fairness in autonomous racing and provides a foundation for developing ethical and safe AI-driven racing systems.

Tutian Tang, Xingyu Ji, Wanli Xing et al.

While Vision-Language-Action (VLA) models have demonstrated remarkable success in robotic manipulation, their application has largely been confined to low-degree-of-freedom end-effectors performing simple, vision-guided pick-and-place tasks. Extending these models to human-like, bimanual dexterous manipulation-specifically contact-rich in-hand operations-introduces critical challenges in high-fidelity data acquisition, multi-skill learning, and multimodal sensory fusion. In this paper, we propose an integrated framework to address these bottlenecks, built upon two components. First, we introduce IMCopilot (In-hand Manipulation Copilot), a suite of reinforcement learning-trained atomic skills that plays a dual role: it acts as a shared-autonomy assistant to simplify teleoperation data collection, and it serves as a callable low-level execution primitive for the VLA. Second, we present MoDE-VLA (Mixture-of-Dexterous-Experts VLA), an architecture that seamlessly integrates heterogeneous force and tactile modalities into a pretrained VLA backbone. By utilizing a residual injection mechanism, MoDE-VLA enables contact-aware refinement without degrading the model's pretrained knowledge. We validate our approach on four tasks of escalating complexity, demonstrating doubled success rate improvement over the baseline in dexterous contact-rich tasks.