arXiv subjects

Shuai Zhang

Publications and source records attributed to Shuai Zhang.

At least 73 records · Page 4Linked to original sources

arXiv2026

General ab initio framework for electronic-order-induced lattice-dynamics symmetry breaking

Conventional \textit{ab initio} approaches are unable to describe phonon time-reversal symmetry ($\mathcal{T}$) breaking. Here, we develop an \textit{ab initio} framework, grounded in molecular Berry curvature (MBC) theory, that captures electronic-order-driven symmetry breaking in lattice dynamics. Using Co$_3$Sn$_2$S$_2$ as a model system, our \textit{ab initio} framework yields phonon spectra that break both $\mathcal{T}$ and mirror symmetries, quantitatively reproduce the observed phonon splittings observed in experiments, and reveal distinct microscopic origins for the $E_g$ and $E_u$ modes: $E_g$ splitting is governed by MBC and is accurately captured by our algorithm, whereas $E_u$ splitting is enhanced by the Fano resonance and matches the experimental data once the Fano-factor correction is included. Leveraging this algorithm, we predict several candidate materials with nonzero electronic-order-driven symmetry breaking in lattice dynamics, establishing a first-principles route to understand electron-phonon coupling, phonon magnetism, and related Hall-type lattice responses.

cond-mat.mtrl-sci↗

arXiv2026

EasyUUV: An LLM-Enhanced Universal and Lightweight Sim-to-Real Reinforcement Learning Framework for UUV Attitude Control

Despite recent advances in Unmanned Underwater Vehicle (UUV) attitude control, existing methods still struggle with generalizability, robustness to real-world disturbances, and efficient deployment. To address the above challenges, this paper presents EasyUUV, a Large Language Model (LLM)-enhanced, platform-agnostic, and lightweight simulation-to-reality reinforcement learning (RL) framework for robust attitude control of UUVs. EasyUUV combines parallelized RL training with a hybrid control architecture, where a learned policy outputs high-level attitude corrections executed by an adaptive S-Surface controller. A multimodal LLM is further integrated to adaptively tune controller parameters at runtime using visual and textual feedback, enabling training-free adaptation to unmodeled dynamics. Also, we have developed a low-cost 6-DoF UUV platform and applied an RL policy trained through efficient parallelized simulation. Extensive simulation and real-world experiments validate the effectiveness and adaptive performance of EasyUUV in achieving robust and adaptive UUV attitude control across diverse underwater conditions. To facilitate reproducibility, the source code, LLM prompts, video, and the supplementary material are provided in the following repositories: Homepage: https://360zmem.github.io/easyuuv/ Video:https://youtu.be/m2yLQzxiIL Supplementary Material: https://drive.google.com/file/d/1ImMiyGIPoPyj2ATnXQwbiuOroTSZ2SSs

cs.RO↗

arXiv2026

Atlas: Orchestrating Heterogeneous Models and Tools for Multi-Domain Complex Reasoning

The integration of large language models (LLMs) with external tools has significantly expanded the capabilities of AI agents. However, as the diversity of both LLMs and tools increases, selecting the optimal model-tool combination becomes a high-dimensional optimization challenge. Existing approaches often rely on a single model or fixed tool-calling logic, failing to exploit the performance variations across heterogeneous model-tool pairs. In this paper, we present ATLAS (Adaptive Tool-LLM Alignment and Synergistic Invocation), a dual-path framework for dynamic tool usage in cross-domain complex reasoning. ATLAS operates via a dual-path approach: (1) \textbf{training-free cluster-based routing} that exploits empirical priors for domain-specific alignment, and (2) \textbf{RL-based multi-step routing} that explores autonomous trajectories for out-of-distribution generalization. Extensive experiments across 15 benchmarks demonstrate that our method outperforms closed-source models like GPT-4o, surpassing existing routing methods on both in-distribution (+10.1%) and out-of-distribution (+13.1%) tasks. Furthermore, our framework shows significant gains in visual reasoning by orchestrating specialized multi-modal tools.

cs.CL↗

arXiv2026

Long-time asymptotics and invariant manifold for the fractional 2D Navier-Stokes equation

We consider the two-dimensional incompressible Navier-Stokes equations with supercritical fractional dissipation in the vorticity formulation. In self-similar variables, we analyze the linearized operator in weighted spaces, prove a spectral gap, and construct a finite-dimensional local slow invariant manifold for small solutions. As a consequence, solutions are attracted to this manifold and admit an explicit long-time asymptotic expansion determined by the leading eigenmodes; additional moment conditions yield faster decay. We also show Lipschitz dependence of the manifold on the dissipation exponent, and recover the classical Navier-Stokes dynamics in the limit as the exponent approaches one.

math.AP↗

arXiv2026

Holo360D: A Large-Scale Real-World Dataset with Continuous Trajectories for Advancing Panoramic 3D Reconstruction and Beyond

While feed-forward 3D reconstruction models have advanced rapidly, they still exhibit degraded performance on panoramas due to spherical distortions. Moreover, existing panoramic 3D datasets are predominantly collected with 360 cameras fixed at discrete locations, resulting in discontinuous trajectories. These limitations critically hinder the development of panoramic feed-forward 3D reconstruction, especially for the multi-view setting. In this paper, we present Holo360D, a comprehensive dataset containing 109,495 panoramas paired with registered point clouds, meshes, and aligned camera poses. To our knowledge, Holo360D is the first large-scale dataset that provides continuous panoramic sequences with accurately aligned high-completeness depth maps. The raw data are initially collected using a 3D laser scanner coupled with a 360 camera. Subsequently, the raw data are processed with both online and offline SLAM systems. Furthermore, to enhance the 3D data quality, a post-processing pipeline tailored for the 360 dataset is proposed, including geometry denoising, mesh hole filling, and region-specific remeshing. Finally, we establish a new benchmark by fine-tuning 3D reconstruction models on Holo360D, providing key insights into effective fine-tuning strategies. Our results demonstrate that Holo360D delivers superior training signals and provides a comprehensive benchmark for advancing panoramic 3D reconstruction models. Datasets and Code will be made publicly available.

cs.CV↗

arXiv2026

TransportBench: A Comprehensive Benchmark for Non-Equilibrium Flow Transport

Scientific machine learning models, as versatile tools for numerical simulation and analysis, are increasingly transforming the landscape of fluid mechanics research. However, existing datasets and benchmarks are primarily limited to continuum fluids and provide limited support for non-equilibrium transport phenomena. To address this gap, we present TransportBench, a high-fidelity dataset and standardized benchmark for non-equilibrium flow transport, designed to reveal the strengths and limitations of neural network models across diverse flow regimes. Specifically, the dataset encompasses a broad physical spectrum, covering continuum and rarefied regimes, low-speed and hypersonic flows, inert and chemically reactive gases, and both translational and internal-energy non-equilibrium effects. Built upon this dataset, we systematically benchmark representative neural architectures using unified evaluation protocols to probe key challenges in learning non-equilibrium flows, including robustness to shock-dominated discontinuities and multi-scale effects, as well as generalization across geometry and physical parameters. Numerical results demonstrate that model performance exhibits a pronounced dependence upon the specific flow characteristics. No single architecture consistently performs best for all the tasks. Instead, different architectural inductive biases provide distinct advantages in capturing smooth flow fields, shock-induced discontinuities, and high-order non-equilibrium statistics. By jointly providing the non-equilibrium flow dataset and model benchmark, TransportBench offers a new testbed for the development, evaluation, and diagnosis of scientific machine learning methods for fluid transport beyond the Navier-Stokes hydrodynamics. The benchmark datasets and implementation codes are available under the MIT license.

physics.comp-ph↗

arXiv2026

Message Tuning Outshines Graph Prompt Tuning: A Prismatic Space Perspective

Graph Foundation Models (GFMs), built upon the Pre-training and Adaptation paradigm, have emerged as a research hotspot in graph learning. For GNN-based GFMs, graph prompt tuning has become the prevailing adaptation method for downstream tasks. Although recent methods explain why graph prompt tuning works, how to rigorously measure its adaptation capacity remains an open problem. Addressing this problem is critical for understanding the capability limits of graph prompt tuning and for developing more powerful adaptation methods. In this paper, we propose Prismatic Space Theory (PS-Theory), a novel mathematical framework to quantify the capacity of adaptation methods, while focusing on establishing the upper bound for the adaptation capacity of graph prompt tuning. Building upon the proposed PS-Theory, we further introduce Message Tuning for GFMs (MTG), a lightweight approach that injects a small set of learnable message prototypes into each layer of the GNN backbone to adaptively guide message fusion without updating pre-trained weights. Through our PS-Theory, we prove that the adaptation capacity of MTG can exceed the theoretical upper bound of graph prompt tuning. Extensive experiments demonstrate that MTG consistently outperforms graph prompt baselines across diverse benchmark datasets, providing strong empirical support for our theoretical findings.

cs.LG↗

arXiv2026

Mechanochemical Nano-Writing of an Atomically Thin Metal

Mechanical energy accelerates many physicochemical processes, including materials syntheses that are hard to produce with thermal energy alone. However, physical understanding connecting applied mechanical forces with internal stresses and ensuing reaction mechanisms is lacking. Here we demonstrate mechanical force-enabled synthesis and nanoscale patterning to metallize a two-dimensional (2D) material, producing an atomically-thin superconducting material. Localized force applied by atomic force microscope tips to van der Waals (vdW) encapsulated stacks of 2D bilayer MoTe2 and adjacent source Pd guides 2D Pd7MoTe2 growth with 50 nm lateral resolution. Force accelerates reaction kinetics exponentially per Eyring's stress-assisted thermal activation model, reducing synthesis temperatures from ~200 °C to near-room temperature. Finite element simulations, density functional theory, and ab-initio grand canonical Monte Carlo calculations show that tip-induced compression facilitates Pd chemisorption to tensile-strained MoTe2 that converts to uniform Pd7MoTe2. This demonstrates a new, generalizable paradigm for nanoscale synthesis of quantum materials, and high-precision engineering of superconductivity.

cond-mat.mes-hall↗

arXiv2026

Beyond Pairwise Interactions: Equivariant Hypergraph Diffusion for Crystal Structure Prediction

Crystal Structure Prediction (CSP) remains a fundamental challenge with significant implications for materials discovery and the advancement of various scientific disciplines. Recent advances have demonstrated that generative models, particularly diffusion models, are especially promising for CSP. However, traditional graph-based representations, where atomic bonds are modeled as pairwise graph edges, fail to capture the intricate high-order interactions essential for accurately describing crystal structures. To address this limitation, we propose leveraging hypergraphs to represent crystal structures, enabling more expressive modeling of multi-way atomic interactions. Hypergraphs naturally encode complex high-order relationships and respect key symmetries -- such as permutation and periodic translation invariance -- that are crucial for characterizing crystalline materials. Building on this representation, we propose the \textbf{E}quivariant \textbf{H}ypergraph \textbf{Diff}usion Model (\textbf{EH-Diff}), a generative framework designed to exploit the symmetry-preserving properties of hypergraphs. EH-Diff provides an efficient and accurate method for predicting crystal structures, with rigorous theoretical guarantees on invariance preservation. Empirically, we conduct extensive experiments on four benchmark datasets, and the results demonstrate that EH-Diff outperforms state-of-the-art CSP methods even with a single diffusion sample.

cs.CE↗

arXiv2026

Cost-Aware Diffusion Draft Trees for Speculative Decoding

Speculative decoding accelerates inference by having a lightweight drafter propose tokens verified in parallel by the target language model. Block diffusion drafters such as DFlash generate an entire draft block in one pass, yielding per-position marginals; DDTree uses these to build a candidate tree that maximizes expected acceptance length under a fixed node budget. We observe, however, that acceptance length is non-decreasing in budget: it always favors larger trees regardless of verification cost, offering no principled basis for budget selection. We introduce \textbf{CaDDTree} (Cost-aware Diffusion Draft Tree), a method that directly optimizes token throughput (expected tokens generated per unit time) by jointly selecting the tree structure and node budget. We model draft and verification latencies explicitly, show that the throughput objective decomposes into a per-round one-dimensional search over the budget, and prove that under a convex verification cost the throughput function is \emph{unimodal}, enabling an efficient greedy stopping rule. CaDDTree requires no offline budget search, adapting the budget each round from the current per-position distributions and verification cost. Experiments on Qwen3-4B and Qwen3-8B across eight benchmarks spanning reasoning, coding, and instruction-following tasks show that \caDDTree{} matches or surpasses DDTree with oracle budget selection on nearly all tasks.

cs.CL↗

arXiv2026

AbstainGNN: Teaching Graph Neural Networks to Abstain for Graph Classification

Graph classification is a core task in graph data mining with widespread real-world applications. Recent advances in graph neural networks (GNNs) have led to substantial performance improvements for graph classification. However, existing GNNs are typically forced to make predictions even under high uncertainty or unknown conditions, resulting in unreliable decisions that can severely impact downstream tasks, particularly in safety-critical scenarios. To address this critical limitation, we propose AbstainGNN, a novel and theory-driven framework for graph classification with abstention, which enables GNNs to reject uncertain predictions instead of producing incorrect decisions. Specifically, AbstainGNN explicitly models both the predictive function and the abstention function, allowing for effective utilization of graph structural information. Moreover, unlike existing heuristic abstention methods, we theoretically characterize the trade-off between classification errors and rejection costs from a PAC-Bayesian generalization perspective, and derive a unified learning objective for model optimization. Guided by this theoretical insight, we further develop an efficient two-stage training strategy consisting of predictive function warm-start and abstention function calibration. Extensive experiments on five benchmark datasets show that AbstainGNN outperforms existing abstention methods, achieving superior classification performance under the same rejection rates.

cs.LG↗

arXiv2026

Cloaking of Arbitrarily Shaped Large-Scale Objects Through the Injection of Electromagnetic Invisibility Genes

Full-space electromagnetic invisibility mainly includes light-bending and scattering-cancellation cloaking. Light-bending cloaking causes double-blind phenomenon and is incompatible with sensing, while scattering-cancellation cloaking allows signal interaction and is more suitable for sensors and communication systems. However, traditional scattering-cancellation cloaking depends highly on target shape and size, making it difficult to realize cloaking for irregular, inhomogeneous and electrically large objects. To solve these problems, this work proposes an electromagnetic invisibility gene injection strategy inspired by biological camouflage. Objects are decomposed into subwavelength units, and customized invisibility genes are injected into each unit according to electromagnetic parameters to achieve overall scattering cancellation. Simulations and microwave experiments verify that this method can realize efficient cloaking for objects with arbitrary shapes, dielectric constants from 2 to 10, and different unit morphologies. This strategy breaks the limits of traditional cloaking and provides a universal, flexible scheme for practical applications such as antenna supports and electromagnetic transparent covers.

physics.optics↗

arXiv2026

Accelerating Robot Path Planning via Connectivity-Preserving Region Proposal Network

Mobile robot path planning methods are often constrained by vast search spaces, resulting in latency in samplingbased algorithms. Learning-based approaches frequently suffer from local region fragmentation and global topological inconsistency. To tackle the problem, we present the Connectivity- Preserving Region Proposal Network (CP-RPN), a segmentationguided model designed to predict compact and topologically connected candidate regions, significantly compressing the search space. Specifically, we design a segmentation model that leverages a Deformable Attention Transformer (DAT) to capture long-range dependencies for global connectivity, with a Deconvolutional decoder to preserve fine-grained spatial details. To guarantee the connectivity of the predicted mask, we design a composite loss function that combines Cross-Entropy loss for pixelwise supervision, a Connectivity-Aware loss to enhance local coherence, and a Topological Continuity loss based on persistent homology to enforce global connectivity. Building on these highconnectivity corridor-like regions, the Voronoi diagram is used to plan the path, backed by a local A* fallback mechanism to ensure robustness. Experimental results demonstrate that CPRPN reduces the candidate region size by over 60.13% compared to the MPT baseline and achieves deterministic low-latency planning (avg. 0.11s) with a 99.60% success rate, outperforming traditional sampling-based algorithms in stability.

cs.RO↗

arXiv2026

Direct simulations of H-He mixtures at planetary interior conditions: demixing, insulator-metal transition and miscibility boundaries

Accurate knowledge of the electrical and thermal conductivities and structural properties of hydrogen-helium mixtures under thermodynamic conditions within and beyond the immiscibility range is very important to predict the thermal evolution and internal structure of gas giant planets like Jupiter and Saturn. Here, we propose a novel method to determine the immiscibility boundary accurately without the need for free energy calculations, while providing consistent insights into structural and transport properties of mixtures. We show with direct large-scale ab initio simulations that the insulator-metal transition (IMT) of the hydrogen subsystem is strongly affected by an admixture with a small fraction of helium and occurs at temperatures significantly higher than those of pure hydrogen. At pressures below 150 GPa, the IMT boundary is not related anymore to the H2 subsystem dissociation, the system remains insulating even after the full dissociation of H2 molecules and its transition to an H-He mixture. The offset of the IMT in the H-He mixture relative to the dissociation region in the hydrogen subsystem and the significant reduction of static electrical and thermal conductivity by a factor between two and a few thousand relative to pure hydrogen found in mixtures have consequences for Jupiter and Saturn's thermal evolution, internal structure, and dynamo action, affecting a large fraction of the interior of both planets.

astro-ph.EP↗

arXiv2026

Emergence of Triplet Superconductivity from Cavity Vacuum Fluctuations

Engineering quantum materials with cavity fields has emerged as a powerful route to manipulate phases of quantum matter in solids. Here we demonstrate that cavity vacuum fluctuations alone can drive the emergence of triplet superconductivity in an otherwise singlet superconductor. The vacuum field renormalizes the electronic band structure in a polarization dependent manner, reshaping the Fermi surface and altering the competition among symmetry allowed pairing channels. As a result, multiple superconducting phases arise from the cavity vacuum fluctuations. Above a critical light matter coupling, the leading instability switches from singlet to triplet pairing, yielding a superconducting state absent in the bare material. This vacuum induced symmetry transition produces distinct modifications of the gap structure and low energy quasiparticle spectrum. Our results establish cavity vacuum engineering as a mechanism for generating unconventional superconducting phases and stabilizing triplet states of potential relevance for topological superconductivity.

cond-mat.supr-con↗

arXiv2026

ERFSL: An Efficient Reward Function Searcher via Language Models for Custom-Environment Multi-Objective Optimization (Student Abstract)

We propose ERFSL, an efficient reward function searcher using large language models (LLMs) for custom-environment, multi-objective learning-based methods (LB). ERFSL generates reward components based on explicit user requirements, rectifies them using a reward critic, and iteratively optimizes the weights of these components based on textual context generated by the training log analyzer. Applied to a simulation-based benchmark task, the reward critic corrects reward codes with only one feedback iteration per requirement, and the reward weight initializer acquires diverse reward functions within the Pareto set. Even when a weight is off by a factor of 500, an average of only 5.2 iterations is needed to meet user requirements. The approach works adequately with GPT-4o mini and does not require advanced understanding capabilities.

eess.SY↗

arXiv2026

FLEX-MoE: Federated Mixture-of-Experts with Load-balanced Expert Assignment for Edge Computing

Mixture-of-Experts (MoE) models enable scalable neural networks through conditional computation, offering enhanced effectiveness and efficiency for next-generation wireless communications. However, deploying MoE with federated learning (FL) over wireless and IoT edge networks faces two critical challenges: 1) resource-constrained clients cannot store large AI models with full expert sets, and 2) non-IID data distributions cause severe expert load imbalance that degrades model performance. To this end, we propose FLEX-MoE, a federated MoE framework that jointly optimizes expert assignment and load balancing under limited client capacity. Specifically, our approach introduces client-expert fitness scores that quantify expert suitability for local datasets through training feedback, and employs an optimization-based algorithm to maximize client-expert specialization while enforcing balanced expert utilization system-wide. Unlike greedy methods that focus solely on personalization while ignoring load imbalance, FLEX-MoE addresses expert utilization skew, which is particularly severe in heterogeneous edge FL. Our experimental results demonstrate superior accuracy and consistently balanced expert utilization across diverse resource-constrained scenarios for edge computing.

cs.LG↗

arXiv2026

Enhancing Information Freshness: An AoI Optimized Markov Decision Process

Ocean exploration utilizing autonomous underwater vehicles (AUVs) via reinforcement learning (RL) has emerged as a significant research focus. However, underwater tasks have mostly failed due to the observation delay caused by information limitation in the information updating networks. In this study, we present an AoI optimized Markov decision process (AoI-MDP) to improve the performance of underwater tasks. Specifically, AoI-MDP models observation delay as timing delay through statistical delay formulation, and includes this delay as a new component in the state space. Additionally, we introduce wait time in the action space, and integrate AoI with reward functions to achieve joint optimization of information freshness and decision-making for AUVs leveraging RL for training. Finally, we apply this approach to the multi-AUV data collection task scenario as an example. Simulation results highlight the feasibility of AoI-MDP, which effectively minimizes AoI while showcasing superior performance in the task. To accelerate relevant research in this field, we have made the simulation codes available as open-source.

eess.SY↗