Pangu Ultra MoE: How to Train Your Big MoE on Ascend NPUs

Sparse large language models (LLMs) with Mixture of Experts (MoE) and close to a trillion parameters are dominating the realm of most capable language models. However, the massive model scale poses significant challenges for the underlying software and hardware systems. In this paper, we aim to uncover a recipe to harness such scale on Ascend NPUs. The key goals are better usage of the computing resources under the dynamic sparse model structures and materializing the expected performance gain on the actual hardware. To select model configurations suitable for Ascend NPUs without repeatedly running the expensive experiments, we leverage simulation to compare the trade-off of various model hyperparameters. This study led to Pangu Ultra MoE, a sparse LLM with 718 billion parameters, and we conducted experiments on the model to verify the simulation results. On the system side, we dig into Expert Parallelism to optimize the communication between NPU devices to reduce the synchronization overhead. We also optimize the memory efficiency within the devices to further reduce the parameter and activation management overhead. In the end, we achieve an MFU of 30.0% when training Pangu Ultra MoE, with performance comparable to that of DeepSeek R1, on 6K Ascend NPUs, and demonstrate that the Ascend system is capable of harnessing all the training stages of the state-of-the-art language models. Extensive experiments indicate that our recipe can lead to efficient training of large-scale sparse language models with MoE. We also study the behaviors of such models for future reference.

Multi-segment Soft Robot Control via Deep Koopman-based Model Predictive Control

Soft robots, compared to regular rigid robots, as their multiple segments with soft materials bring flexibility and compliance, have the advantages of safe interaction and dexterous operation in the environment. However, due to its characteristics of high dimensional, nonlinearity, time-varying nature, and infinite degree of freedom, it has been challenges in achieving precise and dynamic control such as trajectory tracking and position reaching. To address these challenges, we propose a framework of Deep Koopman-based Model Predictive Control (DK-MPC) for handling multi-segment soft robots. We first employ a deep learning approach with sampling data to approximate the Koopman operator, which therefore linearizes the high-dimensional nonlinear dynamics of the soft robots into a finite-dimensional linear representation. Secondly, this linearized model is utilized within a model predictive control framework to compute optimal control inputs that minimize the tracking error between the desired and actual state trajectories. The real-world experiments on the soft robot "Chordata" demonstrate that DK-MPC could achieve high-precision control, showing the potential of DK-MPC for future applications to soft robots.

A General Infrastructure and Workflow for Quadrotor Deep Reinforcement Learning and Reality Deployment

Deploying robot learning methods to a quadrotor in unstructured outdoor environments is an exciting task. Quadrotors operating in real-world environments by learning-based methods encounter several challenges: a large amount of simulator generated data required for training, strict demands for real-time processing onboard, and the sim-to-real gap caused by dynamic and noisy conditions. Current works have made a great breakthrough in applying learning-based methods to end-to-end control of quadrotors, but rarely mention the infrastructure system training from scratch and deploying to reality, which makes it difficult to reproduce methods and applications. To bridge this gap, we propose a platform that enables the seamless transfer of end-to-end deep reinforcement learning (DRL) policies. We integrate the training environment, flight dynamics control, DRL algorithms, the MAVROS middleware stack, and hardware into a comprehensive workflow and architecture that enables quadrotors' policies to be trained from scratch to real-world deployment in several minutes. Our platform provides rich types of environments including hovering, dynamic obstacle avoidance, trajectory tracking, balloon hitting, and planning in unknown environments, as a physical experiment benchmark. Through extensive empirical validation, we demonstrate the efficiency of proposed sim-to-real platform, and robust outdoor flight performance under real-world perturbations. Details can be found from our website https://emnavi.tech/AirGym/.

Step-Audio: Unified Understanding and Generation in Intelligent Speech Interaction

Real-time speech interaction, serving as a fundamental interface for human-machine collaboration, holds immense potential. However, current open-source models face limitations such as high costs in voice data collection, weakness in dynamic control, and limited intelligence. To address these challenges, this paper introduces Step-Audio, the first production-ready open-source solution. Key contributions include: 1) a 130B-parameter unified speech-text multi-modal model that achieves unified understanding and generation, with the Step-Audio-Chat version open-sourced; 2) a generative speech data engine that establishes an affordable voice cloning framework and produces the open-sourced lightweight Step-Audio-TTS-3B model through distillation; 3) an instruction-driven fine control system enabling dynamic adjustments across dialects, emotions, singing, and RAP; 4) an enhanced cognitive architecture augmented with tool calling and role-playing abilities to manage complex tasks effectively. Based on our new StepEval-Audio-360 evaluation benchmark, Step-Audio achieves state-of-the-art performance in human evaluations, especially in terms of instruction following. On open-source benchmarks like LLaMA Question, shows 9.3% average performance improvement, demonstrating our commitment to advancing the development of open-source multi-modal language technologies. Our code and models are available at https://github.com/stepfun-ai/Step-Audio.

Multilayered Intelligent Reflecting Surface for Long-Range Underwater Acoustic Communication

This article introduces a multilayered acoustic reconfigurable intelligent surface (ML-ARIS) architecture designed for the next generation of underwater communications. ML-ARIS incorporates multiple layers of piezoelectric material in each acoustic reflector, with the load impedance of each layer independently adjustable via a control circuit. This design increases the flexibility in generating reflected signals with desired amplitudes and orthogonal phases, enabling passive in-phase and quadrature (IQ) modulation using a single acoustic reflector. Such a feature enables precise beam steering, enhancing sound levels in targeted directions while minimizing interference in surrounding environments. Extensive simulations and tank experiments were conducted to verify the feasibility of ML-ARIS. The experimental results indicate that implementing IQ modulation with a multilayer structure is indeed practical in real-world scenarios, making it possible to use a single reflection unit to generate reflected waves with high-resolution amplitudes and phases.

Underwater Acoustic Reconfigurable Intelligent Surfaces: from Principle to Practice

This article explores the potential of underwater acoustic reconfigurable intelligent surfaces (UA-RIS) for facilitating long-range and eco-friendly communication in marine environments. Unlike radio frequency-based RIS (RF-RIS), which have been extensively investigated in terrestrial contexts, UA-RIS is an emerging field of study. The distinct characteristics of acoustic waves, including their slow propagation speed and potential for noise pollution affecting marine life, necessitate a fundamentally different approach to the architecture and design principles of UA-RIS compared to RF-RIS. Currently, there is a scarcity of real systems and experimental data to validate the feasibility of UA-RIS in practical applications. To fill this gap, this article presents field tests conducted with a prototype UA-RIS consisting of 24 acoustic elements. The results demonstrate that the developed prototype can effectively reflect acoustic waves to any specified directions through passive beamforming, thereby substantially extending the range and data rate of underwater communication systems.

Medical Multimodal Foundation Models in Clinical Diagnosis and Treatment: Applications, Challenges, and Future Directions

Recent advancements in deep learning have significantly revolutionized the field of clinical diagnosis and treatment, offering novel approaches to improve diagnostic precision and treatment efficacy across diverse clinical domains, thus driving the pursuit of precision medicine. The growing availability of multi-organ and multimodal datasets has accelerated the development of large-scale Medical Multimodal Foundation Models (MMFMs). These models, known for their strong generalization capabilities and rich representational power, are increasingly being adapted to address a wide range of clinical tasks, from early diagnosis to personalized treatment strategies. This review offers a comprehensive analysis of recent developments in MMFMs, focusing on three key aspects: datasets, model architectures, and clinical applications. We also explore the challenges and opportunities in optimizing multimodal representations and discuss how these advancements are shaping the future of healthcare by enabling improved patient outcomes and more efficient clinical workflows.

Bidirectional-Reachable Hierarchical Reinforcement Learning with Mutually Responsive Policies

Hierarchical reinforcement learning (HRL) addresses complex long-horizon tasks by skillfully decomposing them into subgoals. Therefore, the effectiveness of HRL is greatly influenced by subgoal reachability. Typical HRL methods only consider subgoal reachability from the unilateral level, where a dominant level enforces compliance to the subordinate level. However, we observe that when the dominant level becomes trapped in local exploration or generates unattainable subgoals, the subordinate level is negatively affected and cannot follow the dominant level's actions. This can potentially make both levels stuck in local optima, ultimately hindering subsequent subgoal reachability. Allowing real-time bilateral information sharing and error correction would be a natural cure for this issue, which motivates us to propose a mutual response mechanism. Based on this, we propose the Bidirectional-reachable Hierarchical Policy Optimization (BrHPO)--a simple yet effective algorithm that also enjoys computation efficiency. Experiment results on a variety of long-horizon tasks showcase that BrHPO outperforms other state-of-the-art HRL baselines, coupled with a significantly higher exploration efficiency and robustness.

SAM-VMNet: Deep Neural Networks For Coronary Angiography Vessel Segmentation

Coronary artery disease (CAD) is one of the most prevalent diseases in the cardiovascular field and one of the major contributors to death worldwide. Computed Tomography Angiography (CTA) images are regarded as the authoritative standard for the diagnosis of coronary artery disease, and by performing vessel segmentation and stenosis detection on CTA images, physicians are able to diagnose coronary artery disease more accurately. In order to combine the advantages of both the base model and the domain-specific model, and to achieve high-precision and fully-automatic segmentation and detection with a limited number of training samples, we propose a novel architecture, SAM-VMNet, which combines the powerful feature extraction capability of MedSAM with the advantage of the linear complexity of the visual state-space model of VM-UNet, giving it faster inferences than Vision Transformer with faster inference speed and stronger data processing capability, achieving higher segmentation accuracy and stability for CTA images. Experimental results show that the SAM-VMNet architecture performs excellently in the CTA image segmentation task, with a segmentation accuracy of up to 98.32% and a sensitivity of up to 99.33%, which is significantly better than other existing models and has stronger domain adaptability. Comprehensive evaluation of the CTA image segmentation task shows that SAM-VMNet accurately extracts the vascular trunks and capillaries, demonstrating its great potential and wide range of application scenarios for the vascular segmentation task, and also laying a solid foundation for further stenosis detection.

OMPO: A Unified Framework for RL under Policy and Dynamics Shifts

Training reinforcement learning policies using environment interaction data collected from varying policies or dynamics presents a fundamental challenge. Existing works often overlook the distribution discrepancies induced by policy or dynamics shifts, or rely on specialized algorithms with task priors, thus often resulting in suboptimal policy performances and high learning variances. In this paper, we identify a unified strategy for online RL policy learning under diverse settings of policy and dynamics shifts: transition occupancy matching. In light of this, we introduce a surrogate policy learning objective by considering the transition occupancy discrepancies and then cast it into a tractable min-max optimization problem through dual reformulation. Our method, dubbed Occupancy-Matching Policy Optimization (OMPO), features a specialized actor-critic structure equipped with a distribution discriminator and a small-size local buffer. We conduct extensive experiments based on the OpenAI Gym, Meta-World, and Panda Robots environments, encompassing policy shifts under stationary and nonstationary dynamics, as well as domain adaption. The results demonstrate that OMPO outperforms the specialized baselines from different categories in all settings. We also find that OMPO exhibits particularly strong performance when combined with domain randomization, highlighting its potential in RL-based robotics applications