Xiang Zhu

Xiang Zhu, Yichen Liu, Hezhong Li et al.

Recent robot learning methods commonly rely on imitation learning from massive robotic dataset collected with teleoperation. When facing a new task, such methods generally require collecting a set of new teleoperation data and finetuning the policy. Furthermore, the teleoperation data collection pipeline is also tedious and expensive. Instead, human is able to efficiently learn new tasks by just watching others do. In this paper, we introduce a novel two-stage framework that utilizes human demonstrations to learn a generalizable robot policy. Such policy can directly take human demonstration video as a prompt and perform new tasks without any new teleoperation data and model finetuning at all. In the first stage, we train video generation model that captures a joint representation for both the human and robot demonstration video data using cross-prediction. In the second stage, we fuse the learned representation with a shared action space between human and robot using a novel prototypical contrastive loss. Empirical evaluations on real-world dexterous manipulation tasks show the effectiveness and generalization capabilities of our proposed method.

Sijie Shen, Xiang Zhu, Yihong Dong et al. · pku

Code generation aims to generate a code snippet automatically from natural language descriptions. Generally, the mainstream code generation methods rely on a large amount of paired training data, including both the natural language description and the code. However, in some domain-specific scenarios, building such a large paired corpus for code generation is difficult because there is no directly available pairing data, and a lot of effort is required to manually write the code descriptions to construct a high-quality training dataset. Due to the limited training data, the generation model cannot be well trained and is likely to be overfitting, making the model's performance unsatisfactory for real-world use. To this end, in this paper, we propose a task augmentation method that incorporates domain knowledge into code generation models through auxiliary tasks and a Subtoken-TranX model by extending the original TranX model to support subtoken-level code generation. To verify our proposed approach, we collect a real-world code generation dataset and conduct experiments on it. Our experimental results demonstrate that the subtoken-level TranX model outperforms the original TranX model and the Transformer model on our dataset, and the exact match accuracy of Subtoken-TranX improves significantly by 12.75% with the help of our task augmentation method. The model performance on several code categories has satisfied the requirements for application in industrial systems. Our proposed approach has been adopted by Alibaba's BizCook platform. To the best of our knowledge, this is the first domain code generation system adopted in industrial development environments.

Xiang Zhu, Puzhen Yuan, Yichen Liu et al.

Learning generalizable vision-language-action (VLA) models from large-scale human videos is promising but challenging due to cross-embodiment discrepancies in both visual observations and executable actions. While latent action models reduce the action execution gap by learning action abstractions, they still rely on visual features. Thus, misaligned human and robot visual representations can lead to inconsistencies in policy inputs and induce domain-dependent latent actions, hindering effective co-training with human videos. To address this, we propose HARP, a human-robot aligned representation learning framework for more effective VLA pretraining from human videos. Specifically, HARP uses limited paired human-robot demonstrations as cross-embodiment bridges and abundant unpaired human and robot videos as a scalable dynamics supervision data source. It trains a robot-adapted visual encoder and a latent action model with manipulation-centric auxiliary cues and a source-relative pair-discriminative alignment loss, which adapts robot representations toward human semantics while preserving pair-level discrimination. The learned aligned vision encoder and latent action model provide a unified vision and action representation for VLA-style policy learning, where human and robot videos provide vision-language-to-latent-action supervision and a lightweight robot action head grounds latent actions into executable commands. Experiments on feature visualization, simulation, and realworld manipulation show improved human-robot alignment and downstream policy performance, achieving 4.481 average length on CALVIN ABC$\rightarrow$D and a 7.1\% realworld success rate gain over the strongest baseline.

Xinyang Gu, Yen-Jen Wang, Xiang Zhu et al.

Humanoid robots, with their human-like skeletal structure, are especially suited for tasks in human-centric environments. However, this structure is accompanied by additional challenges in locomotion controller design, especially in complex real-world environments. As a result, existing humanoid robots are limited to relatively simple terrains, either with model-based control or model-free reinforcement learning. In this work, we introduce Denoising World Model Learning (DWL), an end-to-end reinforcement learning framework for humanoid locomotion control, which demonstrates the world's first humanoid robot to master real-world challenging terrains such as snowy and inclined land in the wild, up and down stairs, and extremely uneven terrains. All scenarios run the same learned neural network with zero-shot sim-to-real transfer, indicating the superior robustness and generalization capability of the proposed method.

Xiang Zhu, Shucheng Kang, Jianyu Chen

Reinforcement learning shows great potential to solve complex contact-rich robot manipulation tasks. However, the safety of using RL in the real world is a crucial problem, since unexpected dangerous collisions might happen when the RL policy is imperfect during training or in unseen scenarios. In this paper, we propose a contact-safe reinforcement learning framework for contact-rich robot manipulation, which maintains safety in both the task space and joint space. When the RL policy causes unexpected collisions between the robot arm and the environment, our framework is able to immediately detect the collision and ensure the contact force to be small. Furthermore, the end-effector is enforced to perform contact-rich tasks compliantly, while keeping robust to external disturbances. We train the RL policy in simulation and transfer it to the real robot. Real world experiments on robot wiping tasks show that our method is able to keep the contact force small both in task space and joint space even when the policy is under unseen scenario with unexpected collision, while rejecting the disturbances on the main task.

Wenhao Tang, Fengtao Zhou, Sheng Huang et al.

Multiple instance learning (MIL) is the most widely used framework in computational pathology, encompassing sub-typing, diagnosis, prognosis, and more. However, the existing MIL paradigm typically requires an offline instance feature extractor, such as a pre-trained ResNet or a foundation model. This approach lacks the capability for feature fine-tuning within the specific downstream tasks, limiting its adaptability and performance. To address this issue, we propose a Re-embedded Regional Transformer (R$^2$T) for re-embedding the instance features online, which captures fine-grained local features and establishes connections across different regions. Unlike existing works that focus on pre-training powerful feature extractor or designing sophisticated instance aggregator, R$^2$T is tailored to re-embed instance features online. It serves as a portable module that can seamlessly integrate into mainstream MIL models. Extensive experimental results on common computational pathology tasks validate that: 1) feature re-embedding improves the performance of MIL models based on ResNet-50 features to the level of foundation model features, and further enhances the performance of foundation model features; 2) the R$^2$T can introduce more significant performance improvements to various MIL models; 3) R$^2$T-MIL, as an R$^2$T-enhanced AB-MIL, outperforms other latest methods by a large margin.The code is available at: https://github.com/DearCaat/RRT-MIL.

Zelin Sun, Shanshan Jiang, Xiaoyu Peng et al.

This paper proposes a decentralized method for regional pole placement, or $\mathcal{D}$-stability, in linearized networked systems. Existing LMI-based methods are hindered by confidentiality concerns regarding proprietary subsystem models and the absence of communication infrastructures. To overcome these barriers, we map the target region $\mathcal{D}$ of pole placement to an auxiliary left-half plane and introduce positive functions to handle the resulting complex-coefficient dynamics. We prove that $\mathcal{D}$-stability is guaranteed via local frequency-domain criteria without requiring shared subsystem models or inter-subsystem communication. This method is then tailored to DC microgrids, where a loop transformation is utilized to reallocate the burden of stability certification, deriving a broadcastable grid code for decentralized parameter synthesis. Numerical examples verify the efficacy of the proposed method.

Jianke Zhang, Yanjiang Guo, Yucheng Hu et al.

Recent advancements in Vision-Language-Action (VLA) models have leveraged pre-trained Vision-Language Models (VLMs) to improve the generalization capabilities. VLMs, typically pre-trained on vision-language understanding tasks, provide rich semantic knowledge and reasoning abilities. However, prior research has shown that VLMs often focus on high-level semantic content and neglect low-level features, limiting their ability to capture detailed spatial information and understand physical dynamics. These aspects, which are crucial for embodied control tasks, remain underexplored in existing pre-training paradigms. In this paper, we investigate the training paradigm for VLAs, and introduce \textbf{UP-VLA}, a \textbf{U}nified VLA model training with both multi-modal \textbf{U}nderstanding and future \textbf{P}rediction objectives, enhancing both high-level semantic comprehension and low-level spatial understanding. Experimental results show that UP-VLA achieves a 33% improvement on the Calvin ABC-D benchmark compared to the previous state-of-the-art method. Additionally, UP-VLA demonstrates improved success rates in real-world manipulation tasks, particularly those requiring precise spatial information.

Weifeng Xu, Xiang Zhu, Xiaoyong Li

The conditional diffusion model (CDM) enhances the standard diffusion model by providing more control, improving the quality and relevance of the outputs, and making the model adaptable to a wider range of complex tasks. However, inaccurate conditional inputs in the inverse process of CDM can easily lead to generating fixed errors in the neural network, which diminishes the adaptability of a well-trained model. The existing methods like data augmentation, adversarial training, robust optimization can improve the robustness, while they often face challenges such as high computational complexity, limited applicability to unknown perturbations, and increased training difficulty. In this paper, we propose a lightweight solution, the Robust Conditional Diffusion Model (RCDM), based on control theory to dynamically reduce the impact of noise and significantly enhance the model's robustness. RCDM leverages the collaborative interaction between two neural networks, along with optimal control strategies derived from control theory, to optimize the weights of two networks during the sampling process. Unlike conventional techniques, RCDM establishes a mathematical relationship between fixed errors and the weights of the two neural networks without incurring additional computational overhead. Extensive experiments were conducted on MNIST and CIFAR-10 datasets, and the results demonstrate the effectiveness and adaptability of our proposed model.

Qian Wang, Zhihao Zhan, Jialei He et al.

Highly accurate geometric precision and dense image features characterize True Digital Orthophoto Maps (TDOMs), which are in great demand for applications such as urban planning, infrastructure management, and environmental monitoring.Traditional TDOM generation methods need sophisticated processes, such as Digital Surface Models (DSM) and occlusion detection, which are computationally expensive and prone to errors.This work presents an alternative technique rooted in 2D Gaussian Splatting (2DGS), free of explicit DSM and occlusion detection. With depth map generation, spatial information for every pixel within the TDOM is retrieved and can reconstruct the scene with high precision. Divide-and-conquer strategy achieves excellent GS training and rendering with high-resolution TDOMs at a lower resource cost, which preserves higher quality of rendering on complex terrain and thin structure without a decrease in efficiency. Experimental results demonstrate the efficiency of large-scale scene reconstruction and high-precision terrain modeling. This approach provides accurate spatial data, which assists users in better planning and decision-making based on maps.

Wang Pang, Zhihao Zhan, Xiang Zhu et al.

Most motion deblurring algorithms rely on spatial-domain convolution models, which struggle with the complex, non-linear blur arising from camera shake and object motion. In contrast, we propose a novel single-image deblurring approach that treats motion blur as a temporal averaging phenomenon. Our core innovation lies in leveraging a pre-trained video diffusion transformer model to capture diverse motion dynamics within a latent space. It sidesteps explicit kernel estimation and effectively accommodates diverse motion patterns. We implement the algorithm within a diffusion-based inverse problem framework. Empirical results on synthetic and real-world datasets demonstrate that our method outperforms existing techniques in deblurring complex motion blur scenarios. This work paves the way for utilizing powerful video diffusion models to address single-image deblurring challenges.

Jiayi Wan, Xiang Zhu, Fanzhen Liu et al.

Federated learning, as a distributed architecture, shows great promise for applications in Cyber-Physical-Social Systems (CPSS). In order to mitigate the privacy risks inherent in CPSS, the integration of differential privacy with federated learning has attracted considerable attention. Existing research mainly focuses on dynamically adjusting the noise added or discarding certain gradients to mitigate the noise introduced by differential privacy. However, these approaches fail to remove the noise that hinders convergence and correct the gradients affected by the noise, which significantly reduces the accuracy of model classification. To overcome these challenges, this paper proposes a novel framework for differentially private federated learning that balances rigorous privacy guarantees with accuracy by introducing a server-side gradient correction mechanism. Specifically, after clients perform gradient clipping and noise perturbation, our framework detects deviations in the noisy local gradients and employs a projection mechanism to correct them, mitigating the negative impact of noise. Simultaneously, gradient projection promotes the alignment of gradients from different clients and guides the model towards convergence to a global optimum. We evaluate our framework on several benchmark datasets, and the experimental results demonstrate that it achieves state-of-the-art performance under the same privacy budget.

Zhihao Zhan, Wang Pang, Xiang Zhu et al.

In this work, we rethink the approach to video super-resolution by introducing a method based on the Diffusion Posterior Sampling framework, combined with an unconditional video diffusion transformer operating in latent space. The video generation model, a diffusion transformer, functions as a space-time model. We argue that a powerful model, which learns the physics of the real world, can easily handle various kinds of motion patterns as prior knowledge, thus eliminating the need for explicit estimation of optical flows or motion parameters for pixel alignment. Furthermore, a single instance of the proposed video diffusion transformer model can adapt to different sampling conditions without re-training. Empirical results on synthetic and real-world datasets illustrate the feasibility of diffusion-based, alignment-free video super-resolution.

Jialei He, Zhihao Zhan, Zhituo Tu et al.

Rapid generation of large-scale orthoimages from Unmanned Aerial Vehicles (UAVs) has been a long-standing focus of research in the field of aerial mapping. A multi-sensor UAV system, integrating the Global Positioning System (GPS), Inertial Measurement Unit (IMU), 4D millimeter-wave radar and camera, can provide an effective solution to this problem. In this paper, we utilize multi-sensor data to overcome the limitations of conventional orthoimage generation methods in terms of temporal performance, system robustness, and geographic reference accuracy. A prior-pose-optimized feature matching method is introduced to enhance matching speed and accuracy, reducing the number of required features and providing precise references for the Structure from Motion (SfM) process. The proposed method exhibits robustness in low-texture scenes like farmlands, where feature matching is difficult. Experiments show that our approach achieves accurate feature matching orthoimage generation in a short time. The proposed drone system effectively aids in farmland detection and management.

Xiang Zhu, Hossein Talebi, Xinwei Shi et al.

In machine learning based single image super-resolution, the degradation model is embedded in training data generation. However, most existing satellite image super-resolution methods use a simple down-sampling model with a fixed kernel to create training images. These methods work fine on synthetic data, but do not perform well on real satellite images. We propose a realistic training data generation model for commercial satellite imagery products, which includes not only the imaging process on satellites but also the post-process on the ground. We also propose a convolutional neural network optimized for satellite images. Experiments show that the proposed training data generation model is able to improve super-resolution performance on real satellite images.