arXiv subjects

Yu Yang

Publications and source records attributed to Yu Yang.

At least 73 records · Page 4Linked to original sources

arXiv2026

Xiaomi-Robotics-0: An Open-Sourced Vision-Language-Action Model with Real-Time Execution

In this report, we introduce Xiaomi-Robotics-0, an advanced vision-language-action (VLA) model optimized for high performance and fast and smooth real-time execution. The key to our method lies in a carefully designed training recipe and deployment strategy. Xiaomi-Robotics-0 is first pre-trained on large-scale cross-embodiment robot trajectories and vision-language data, endowing it with broad and generalizable action-generation capabilities while avoiding catastrophic forgetting of the visual-semantic knowledge of the underlying pre-trained VLM. During post-training, we propose several techniques for training the VLA model for asynchronous execution to address the inference latency during real-robot rollouts. During deployment, we carefully align the timesteps of consecutive predicted action chunks to ensure continuous and seamless real-time rollouts. We evaluate Xiaomi-Robotics-0 extensively in simulation benchmarks and on two challenging real-robot tasks that require precise and dexterous bimanual manipulation. Results show that our method achieves state-of-the-art performance across all simulation benchmarks. Moreover, Xiaomi-Robotics-0 can roll out fast and smoothly on real robots using a consumer-grade GPU, achieving high success rates and throughput on both real-robot tasks. To facilitate future research, code and model checkpoints are open-sourced at https://xiaomi-robotics-0.github.io

cs.RO↗

arXiv2026

DyMRL: Dynamic Multispace Representation Learning for Multimodal Event Forecasting in Knowledge Graph

Accurate representation of multimodal knowledge is crucial for event forecasting in real-world scenarios. However, existing studies have largely focused on static settings, overlooking the dynamic acquisition and fusion of multimodal knowledge. 1) At the knowledge acquisition level, how to learn time-sensitive information of different modalities, especially the dynamic structural modality. Existing dynamic learning methods are often limited to shallow structures across heterogeneous spaces or simple unispaces, making it difficult to capture deep relation-aware geometric features. 2) At the knowledge fusion level, how to learn evolving multimodal fusion features. Existing knowledge fusion methods based on static coattention struggle to capture the varying historical contributions of different modalities to future events. To this end, we propose DyMRL, a Dynamic Multispace Representation Learning approach to efficiently acquire and fuse multimodal temporal knowledge. 1) For the former issue, DyMRL integrates time-specific structural features from Euclidean, hyperbolic, and complex spaces into a relational message-passing framework to learn deep representations, reflecting human intelligences in associative thinking, high-order abstracting, and logical reasoning. Pretrained models endow DyMRL with time-sensitive visual and linguistic intelligences. 2) For the latter concern, DyMRL incorporates advanced dual fusion-evolution attention mechanisms that assign dynamic learning emphases equally to different modalities at different timestamps in a symmetric manner. To evaluate DyMRL's event forecasting performance through leveraging its learned multimodal temporal knowledge in history, we construct four multimodal temporal knowledge graph benchmarks. Extensive experiments demonstrate that DyMRL outperforms state-of-the-art dynamic unimodal and static multimodal baseline methods.

cs.LG↗

arXiv2026

MOBODY: Model Based Off-Dynamics Offline Reinforcement Learning

We study off-dynamics offline reinforcement learning, where the goal is to learn a policy from offline source and limited target datasets with mismatched dynamics. Existing methods either penalize the reward or discard source transitions occurring in parts of the transition space with high dynamics shift. As a result, they optimize the policy using data from low-shift regions, limiting exploration of high-reward states in the target domain that do not fall within these regions. Consequently, such methods often fail when the dynamics shift is significant or the optimal trajectories lie outside the low-shift regions. To overcome this limitation, we propose MOBODY, a Model-Based Off-Dynamics Offline RL algorithm that optimizes a policy using learned target dynamics transitions to explore the target domain, rather than only being trained with the low dynamics-shift transitions. For the dynamics learning, built on the observation that achieving the same next state requires taking different actions in different domains, MOBODY employs separate action encoders for each domain to encode different actions to the shared latent space while sharing a unified representation of states and a common transition function. We further introduce a target Q-weighted behavior cloning loss in policy optimization to avoid out-of-distribution actions, which push the policy toward actions with high target-domain Q-values, rather than high source domain Q-values or uniformly imitating all actions in the offline dataset. We evaluate MOBODY on a wide range of MuJoCo and Adroit benchmarks, demonstrating that it outperforms state-of-the-art off-dynamics RL baselines as well as policy learning methods based on different dynamics learning baselines, with especially pronounced improvements in challenging scenarios where existing methods struggle.

cs.LG↗

arXiv2026

Return Augmented Decision Transformer for Off-Dynamics Reinforcement Learning

We study offline off-dynamics reinforcement learning (RL) to utilize data from an easily accessible source domain to enhance policy learning in a target domain with limited data. Our approach centers on return-conditioned supervised learning (RCSL), particularly focusing on Decision Transformer (DT) type frameworks, which can predict actions conditioned on desired return guidance and complete trajectory history. Previous works address the dynamics shift problem by augmenting the reward in the trajectory from the source domain to match the optimal trajectory in the target domain. However, this strategy can not be directly applicable in RCSL owing to (1) the unique form of the RCSL policy class, which explicitly depends on the return, and (2) the absence of a straightforward representation of the optimal trajectory distribution. We propose the Return Augmented (REAG) method for DT type frameworks, where we augment the return in the source domain by aligning its distribution with that in the target domain. We provide the theoretical analysis demonstrating that the RCSL policy learned from REAG achieves the same level of suboptimality as would be obtained without a dynamics shift. We introduce two practical implementations REAG$_\text{Dara}^{*}$ and REAG$_\text{MV}^{*}$ respectively. Thorough experiments on D4RL datasets and various DT-type baselines demonstrate that our methods consistently enhance the performance of DT type frameworks in off-dynamics RL.

cs.LG↗

arXiv2026

Bridging Policy and Real-World Dynamics: LLM-Augmented Rebalancing for Shared Micromobility Systems

Shared micromobility services such as e-scooters and bikes have become an integral part of urban transportation, yet their efficiency critically depends on effective vehicle rebalancing. Existing methods either optimize for average demand patterns or employ robust optimization and reinforcement learning to handle predefined uncertainties. However, these approaches overlook emergent events (e.g., demand surges, vehicle outages, regulatory interventions) or sacrifice performance in normal conditions. We introduce AMPLIFY, an LLM-augmented policy adaptation framework for shared micromobility rebalancing. The framework combines a baseline rebalancing module with an LLM-based adaptation module that adjusts strategies in real time under emergent scenarios. The adaptation module ingests system context, demand predictions, and baseline strategies, and refines adjustments through self-reflection. Evaluations on real-world e-scooter data from Chicago show that our approach improves demand satisfaction and system revenue compared to baseline policies, highlighting the potential of LLM-driven adaptation as a flexible solution for managing uncertainty in micromobility systems.

cs.LG↗

arXiv2026

Localized Dynamics-Aware Domain Adaption for Off-Dynamics Offline Reinforcement Learning

Off-dynamics offline reinforcement learning (RL) aims to learn a policy for a target domain using limited target data and abundant source data collected under different transition dynamics. Existing methods typically address dynamics mismatch either globally over the state space or via pointwise data filtering; these approaches can miss localized cross-domain similarities or incur high computational cost. We propose Localized Dynamics-Aware Domain Adaptation (LoDADA), which exploits localized dynamics mismatch to better reuse source data. LoDADA clusters transitions from source and target datasets and estimates cluster-level dynamics discrepancy via domain discrimination. Source transitions from clusters with small discrepancy are retained, while those from clusters with large discrepancy are filtered out. This yields a fine-grained and scalable data selection strategy that avoids overly coarse global assumptions and expensive per-sample filtering. We provide theoretical insights and extensive experiments across environments with diverse global and local dynamics shifts. Results show that LoDADA consistently outperforms state-of-the-art off-dynamics offline RL methods by better leveraging localized distribution mismatch.

cs.LG↗

arXiv2026

Efficient and Stable Reinforcement Learning for Diffusion Language Models

Reinforcement Learning (RL) is crucial for unlocking the complex reasoning capabilities of Diffusion-based Large Language Models (dLLMs). However, applying RL to dLLMs faces unique challenges in efficiency and stability. To address these challenges, we propose Spatio-Temporal Pruning (STP), a framework designed to simultaneously improve the efficiency and stability of RL for dLLMs. STP compresses the redundancy in the generative process through: (1) \textit{spatial pruning}, which constrains the exploration space using static priors; and (2) \textit{temporal pruning}, which bypasses redundant late-stage refinement steps. Our theoretical analysis demonstrates that STP strictly reduces the variance of the log-likelihood estimation, thereby ensuring more stable policy updates. Extensive experiments demonstrate that STP surpasses state-of-the-art baselines in both efficiency and accuracy. Our code is available at https://github.com/Lolo1222/STP.

cs.AI↗

arXiv2026

Vision-Centric 4D Occupancy Forecasting and Planning via Implicit Residual World Models

End-to-end autonomous driving systems increasingly rely on vision-centric world models to understand and predict their environment. However, a common ineffectiveness in these models is the full reconstruction of future scenes, which expends significant capacity on redundantly modeling static backgrounds. To address this, we propose IR-WM, an Implicit Residual World Model that focuses on modeling the current state and evolution of the world. IR-WM first establishes a robust bird's-eye-view representation of the current state from the visual observation. It then leverages the BEV features from the previous timestep as a strong temporal prior and predicts only the "residual", i.e., the changes conditioned on the ego-vehicle's actions and scene context. To alleviate error accumulation over time, we further apply an alignment module to calibrate semantic and dynamic misalignments. Moreover, we investigate different forecasting-planning coupling schemes and demonstrate that the implicit future state generated by world models substantially improves planning accuracy. On the nuScenes benchmark, IR-WM achieves top performance in both 4D occupancy forecasting and trajectory planning.

cs.CV↗

arXiv2026

CoBA-RL: Capability-Oriented Budget Allocation for Reinforcement Learning in LLMs

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a key approach for enhancing LLM reasoning. However, standard frameworks like Group Relative Policy Optimization (GRPO) typically employ a uniform rollout budget, leading to resource inefficiency. Moreover, existing adaptive methods often rely on instance-level metrics, such as task pass rates, failing to capture the model's dynamic learning state. To address these limitations, we propose CoBA-RL, a reinforcement learning algorithm designed to adaptively allocate rollout budgets based on the model's evolving capability. Specifically, CoBA-RL utilizes a Capability-Oriented Value function to map tasks to their potential training gains and employs a heap-based greedy strategy to efficiently self-calibrate the distribution of computational resources to samples with high training value. Extensive experiments demonstrate that our approach effectively orchestrates the trade-off between exploration and exploitation, delivering consistent generalization improvements across multiple challenging benchmarks. These findings underscore that quantifying sample training value and optimizing budget allocation are pivotal for advancing LLM post-training efficiency.

cs.LG↗

arXiv2026

ScaleEnv: Scaling Environment Synthesis from Scratch for Generalist Interactive Tool-Use Agent Training

Training generalist agents capable of adapting to diverse scenarios requires interactive environments for self-exploration. However, interactive environments remain critically scarce, and existing synthesis methods suffer from significant limitations regarding environmental diversity and scalability. To address these challenges, we introduce ScaleEnv, a framework that constructs fully interactive environments and verifiable tasks entirely from scratch. Specifically, ScaleEnv ensures environment reliability through procedural testing, and guarantees task completeness and solvability via tool dependency graph expansion and executable action verification. By enabling agents to learn through exploration within ScaleEnv, we demonstrate significant performance improvements on unseen, multi-turn tool-use benchmarks such as $τ^2$-Bench and VitaBench, highlighting strong generalization capabilities. Furthermore, we investigate the relationship between increasing number of domains and model generalization performance, providing empirical evidence that scaling environmental diversity is critical for robust agent learning.

cs.AI↗

arXiv2026

FedRD: Reducing Divergences for Generalized Federated Learning via Heterogeneity-aware Parameter Guidance

Heterogeneous federated learning (HFL) aims to ensure effective and privacy-preserving collaboration among different entities. As newly joined clients require significant adjustments and additional training to align with the existing system, the problem of generalizing federated learning models to unseen clients under heterogeneous data has become progressively crucial. Consequently, we highlight two unsolved challenging issues in federated domain generalization: Optimization Divergence and Performance Divergence. To tackle the above challenges, we propose FedRD, a novel heterogeneity-aware federated learning algorithm that collaboratively utilizes parameter-guided global generalization aggregation and local debiased classification to reduce divergences, aiming to obtain an optimal global model for participating and unseen clients. Extensive experiments on public multi-domain datasets demonstrate that our approach exhibits a substantial performance advantage over competing baselines in addressing this specific problem.

cs.LG↗

arXiv2026

FedCCA: Client-Centric Adaptation against Data Heterogeneity in Federated Learning on IoT Devices

With the rapid development of the Internet of Things (IoT), AI model training on private data such as human sensing data is highly desired. Federated learning (FL) has emerged as a privacy-preserving distributed training framework for this purpuse. However, the data heterogeneity issue among IoT devices can significantly degrade the model performance and convergence speed in FL. Existing approaches limit in fixed client selection and aggregation on cloud server, making the privacy-preserving extraction of client-specific information during local training challenging. To this end, we propose Client-Centric Adaptation federated learning (FedCCA), an algorithm that optimally utilizes client-specific knowledge to learn a unique model for each client through selective adaptation, aiming to alleviate the influence of data heterogeneity. Specifically, FedCCA employs dynamic client selection and adaptive aggregation based on the additional client-specific encoder. To enhance multi-source knowledge transfer, we adopt an attention-based global aggregation strategy. We conducted extensive experiments on diverse datasets to assess the efficacy of FedCCA. The experimental results demonstrate that our approach exhibits a substantial performance advantage over competing baselines in addressing this specific problem.

cs.LG↗

arXiv2026

Representation Learning Enhanced Deep Reinforcement Learning for Optimal Operation of Hydrogen-based Multi-Energy Systems

Hydrogen-based multi-energy systems (HMES) have emerged as a promising low-carbon and energy-efficient solution, as it can enable the coordinated operation of electricity, heating and cooling supply and demand to enhance operational flexibility, improve overall energy efficiency, and increase the share of renewable integration. However, the optimal operation of HMES remains challenging due to the nonlinear and multi-physics coupled dynamics of hydrogen energy storage systems (HESS) (consisting of electrolyters, fuel cells and hydrogen tanks) as well as the presence of multiple uncertainties from supply and demand. To address these challenges, this paper develops a comprehensive operational model for HMES that fully captures the nonlinear dynamics and multi-physics process of HESS. Moreover, we propose an enhanced deep reinforcement learning (DRL) framework by integrating the emerging representation learning techniques, enabling substantially accelerated and improved policy optimization for spatially and temporally coupled complex networked systems, which is not provided by conventional DRL. Experimental studies based on real-world datasets show that the comprehensive model is crucial to ensure the safe and reliable of HESS. In addition, the proposed SR-DRL approaches demonstrate superior convergence rate and performance over conventional DRL counterparts in terms of reducing the operation cost of HMES and handling the system operating constraints. Finally, we provide some insights into the role of representation learning in DRL, speculating that it can reorganize the original state space into a well-structured and cluster-aware geometric representation, thereby smoothing and facilitating the learning process of DRL.

cs.LG↗

arXiv2026

Rate-Distortion-Perception Tradeoff for the Gray-Wyner Problem

We revisit the Gray-Wyner lossy source coding problem and derive the first-order asymptotic optimal rate-distortion-perception region when additional perception constraints are imposed on reproduced source sequences. The optimal trade-off is shown to be governed by a mutual information term involving common information and two conditional rate-distortion-perception functions. The perception constraint requires that the distribution of each reproduced sequence is close to that of the original source sequence, which is motivated by practical applications in image and video compression. Prior studies usually focus on the compression and reconstruction of a single source sequence. In this paper, we generalize the prior results for point-to-point systems to the representative multi-terminal setting of the Gray-Wyner problem with two correlated source sequences. In particular, we integrate the analyses of the distortion and the perception constraints by including the random circular shift operator in the encoding and decoding process directly.

cs.IT↗

arXiv2026

Mechanical Resonator-based Quantum Computing

Hybrid quantum systems combine the unique advantages of different physical platforms with the goal of realizing more powerful and practical quantum information processing devices. Mechanical systems, such as bulk acoustic wave resonators, feature a large number of highly coherent harmonic modes in a compact footprint, which complements the strong nonlinearities and fast operation times of superconducting quantum circuits. Here, we demonstrate an architecture for mechanical resonator-based quantum computing, in which a superconducting qubit is used to perform quantum gates on a collection of mechanical modes. We show the implementation of a universal gate set, composed of single-qubit gates and controlled arbitrary-phase gates, and showcase their use in the quantum Fourier transform and quantum period finding algorithms. These results pave the way toward using mechanical systems to build crucial components for future quantum technologies, such as quantum random-access memories.

quant-ph↗

arXiv2026

Unsteady flow predictions around an obstacle using Geometry-Parameterized Dual-Encoder Physics-Informed Neural Network

Machine learning-based flow field prediction is emerging as a promising alternative to traditional Computational Fluid Dynamics, offering significant computational efficiency advantage. In this work, we propose the Geometry-Parameterized Dual-Encoder Physics-Informed Neural Network (GP-DE-PINN) with a dual-encoder architecture for effective prediction of unsteady flow fields around parameterized geometries. This framework integrates a geometric parameter encoder to map low-dimensional shape parameters to high-dimensional latent features, coupled with a spatiotemporal coordinate encoder, and is trained under the Navier-Stokes equation constraints. Using 2D unsteady flow past petal-shaped cylinders as an example, we evaluate the model's reconstruction performance, generalization capability, and hyperparameter sensitivity. Results demonstrate that the GP-DE-PINN significantly outperforms the PINN with direct geometric input in flow field reconstruction, accurately capturing vortex shedding structures and pressure evolution, while exhibiting superior generalization accuracy on unseen geometric configurations. Furthermore, sensitivity analyses regarding geometric sampling and network width reveal the model's robustness to these hyperparameter variations. These findings illustrate that the proposed framework can serve as a robust and promising framework for predicting unsteady flows around complex geometric obstacles.

physics.flu-dyn↗

arXiv2026

Data-Driven Flow Initialization Framework for CFD Acceleration of Underwater Vehicle in Vertical-Plane Oblique Motion

Accurate prediction of flow fields around underwater vehicles undergoing vertical-plane oblique motions is critical for hydrodynamic analysis, but it often requires computationally expensive CFD simulations. This study proposes a Data-Driven Flow Initialization (DDFI) framework that accelerates CFD simulation by integrating deep neural network (DNN) to predict full-domain flow fields. Using the suboff hull under various inlet velocities and angles of attack as an example, a DNN is trained to predict velocity, pressure, and turbulent quantities based on mesh geometry, operating conditions, and hybrid vectors. The DNN can provide reasonably accurate predictions with a relative error about 3.3%. To enhance numerical accuracy while maintaining physical consistency, the DNN-predicted flow fields are utilized as initial solutions for the CFD solver, achieving up to 3.5-fold and 2.0-fold speedup at residual thresholds of 5*10^(-6)and 5*10^(-8), respectively. This method maintains physical consistency by refining neural network outputs via traditional CFD solvers, balancing computational efficiency and accuracy. Notably, reducing the size of training set does not exert an essential impact on acceleration performance. Besides, this method exhibits cross-mesh generalization capability. In general, this proposed hybrid approach offers a new pathway for high-fidelity and efficient full-domain flow field predictions around complex underwater vehicles.

physics.flu-dyn↗

arXiv2026

Digital Twin AI: Opportunities and Challenges from Large Language Models to World Models

Digital twins, as precise digital representations of physical systems, have evolved from passive simulation tools into intelligent and autonomous entities through the integration of artificial intelligence technologies. This paper presents a unified four-stage framework that systematically characterizes AI integration across the digital twin lifecycle, spanning modeling, mirroring, intervention, and autonomous management. By synthesizing existing technologies and practices, we distill a unified four-stage framework that systematically characterizes how AI methodologies are embedded across the digital twin lifecycle: (1) modeling the physical twin through physics-based and physics-informed AI approaches, (2) mirroring the physical system into a digital twin with real-time synchronization, (3) intervening in the physical twin through predictive modeling, anomaly detection, and optimization strategies, and (4) achieving autonomous management through large language models, foundation models, and intelligent agents. We analyze the synergy between physics-based modeling and data-driven learning, highlighting the shift from traditional numerical solvers to physics-informed and foundation models for physical systems. Furthermore, we examine how generative AI technologies, including large language models and generative world models, transform digital twins into proactive and self-improving cognitive systems capable of reasoning, communication, and creative scenario generation. Through a cross-domain review spanning eleven application domains, including healthcare, aerospace, smart manufacturing, robotics, and smart cities, we identify common challenges related to scalability, explainability, and trustworthiness, and outline directions for responsible AI-driven digital twin systems.

cs.AI↗