arXiv subjects

Weiyi Li

Publications and source records attributed to Weiyi Li.

9 recordsLinked to original sources

arXiv2026

Towards Green Wearable Computing: A Physics-Aware Spiking Neural Network for Energy-Efficient IMU-based Human Activity Recognition

Wearable IMU-based Human Activity Recognition (HAR) relies heavily on Deep Neural Networks (DNNs), which are burdened by immense computational and buffering demands. Their power-hungry floating-point operations and rigid requirement to process complete temporal windows severely cripple battery-constrained edge devices. While Spiking Neural Networks (SNNs) offer extreme event-driven energy efficiency, standard architectures struggle with complex biomechanical topologies and temporal gradient degradation. To bridge this gap, we propose the Physics-Aware Spiking Neural Network (PAS-Net), a fully multiplier-free architecture explicitly tailored for Green HAR. Spatially, an adaptive symmetric topology mixer enforces human-joint physical constraints. Temporally, an $O(1)$-memory causal neuromodulator yields context-aware dynamic threshold neurons, adapting actively to non-stationary movement rhythms. Furthermore, we leverage a temporal spike error objective to unlock a flexible early-exit mechanism for continuous IMU streams. Evaluated across seven diverse datasets, PAS-Net achieves state-of-the-art accuracy while replacing dense operations with sparse 0.1 pJ integer accumulations. Crucially, its confidence-driven early-exit capability drastically reduces dynamic energy consumption by up to 98\%. PAS-Net establishes a robust, ultra-low-power neuromorphic standard for always-on wearable sensing. The source code and pre-trained models are publicly available at https://github.com/zhengnaichuan2022/PAS-Net.git.

cs.LG↗

arXiv2026

S3T-Former: A Purely Spike-Driven State-Space Topology Transformer for Skeleton Action Recognition

Skeleton-based action recognition is crucial for multimedia applications but heavily relies on power-hungry Artificial Neural Networks (ANNs), limiting their deployment on resource-constrained edge devices. Spiking Neural Networks (SNNs) provide an energy-efficient alternative; however, existing spiking models for skeleton data often compromise the intrinsic sparsity of SNNs by resorting to dense matrix aggregations, heavy multimodal fusion modules, or non-sparse frequency domain transformations. Furthermore, they severely suffer from the short-term amnesia of spiking neurons. In this paper, we propose the Spiking State-Space Topology Transformer (S3T-Former), which, to the best of our knowledge, is the first purely spike-driven Transformer architecture specifically designed for energy-efficient skeleton action recognition. Rather than relying on heavy fusion overhead, we formulate a Multi-Stream Anatomical Spiking Embedding (M-ASE) that acts as a generalized kinematic differential operator, elegantly transforming multimodal skeleton features into heterogeneous, highly sparse event streams. To achieve true topological and temporal sparsity, we introduce Lateral Spiking Topology Routing (LSTR) for on-demand conditional spike propagation, and a Spiking State-Space (S3) Engine to systematically capture long-range temporal dynamics without non-sparse spectral workarounds. Extensive experiments on multiple large-scale datasets demonstrate that S3T-Former achieves highly competitive accuracy while theoretically reducing energy consumption compared to classic ANNs, establishing a new state-of-the-art for energy-efficient neuromorphic action recognition.

cs.CV↗

arXiv2025

Signal-SGN++: Topology-Enhanced Time-Frequency Spiking Graph Network for Skeleton-Based Action Recognition

Graph Convolutional Networks (GCNs) demonstrate strong capability in modeling skeletal topology for action recognition, yet their dense floating-point computations incur high energy costs. Spiking Neural Networks (SNNs), characterized by event-driven and sparse activation, offer energy efficiency but remain limited in capturing coupled temporal-frequency and topological dependencies of human motion. To bridge this gap, this article proposes Signal-SGN++, a topology-aware spiking graph framework that integrates structural adaptivity with time-frequency spiking dynamics. The network employs a backbone composed of 1D Spiking Graph Convolution (1D-SGC) and Frequency Spiking Convolution (FSC) for joint spatiotemporal and spectral feature extraction. Within this backbone, a Topology-Shift Self-Attention (TSSA) mechanism is embedded to adaptively route attention across learned skeletal topologies, enhancing graph-level sensitivity without increasing computational complexity. Moreover, an auxiliary Multi-Scale Wavelet Transform Fusion (MWTF) branch decomposes spiking features into multi-resolution temporal-frequency representations, wherein a Topology-Aware Time-Frequency Fusion (TATF) unit incorporates structural priors to preserve topology-consistent spectral fusion. Comprehensive experiments on large-scale benchmarks validate that Signal-SGN++ achieves superior accuracy-efficiency trade-offs, outperforming existing SNN-based methods and achieving competitive results against state-of-the-art GCNs under substantially reduced energy consumption.

cs.CV↗

arXiv2025

Governance-Ready Small Language Models for Medical Imaging: Prompting, Abstention, and PACS Integration

Small Language Models (SLMs) are a practical option for narrow, workflow-relevant medical imaging utilities where privacy, latency, and cost dominate. We present a governance-ready recipe that combines prompt scaffolds, calibrated abstention, and standards-compliant integration into Picture Archiving and Communication Systems (PACS). Our focus is the assistive task of AP/PA view tagging for chest radiographs. Using four deployable SLMs (Qwen2.5-VL, MiniCPM-V, Gemma 7B, LLaVA 7B) on NIH Chest X-ray, we provide illustrative evidence: reflection-oriented prompts benefit lighter models, whereas stronger baselines are less sensitive. Beyond accuracy, we operationalize abstention, expected calibration error, and oversight burden, and we map outputs to DICOM tags, HL7 v2 messages, and FHIR ImagingStudy. The contribution is a prompt-first deployment framework, an operations playbook for calibration, logging, and change management, and a clear pathway from pilot utilities to reader studies without over-claiming clinical validation. We additionally specify a human-factors RACI, stratified calibration for dataset shift, and an auditable evidence pack to support local governance reviews.

cs.CV↗

arXiv2025

Three-dimensional imaging of individual carbon atoms

Carbon is fundamental to science and technology due to its diverse bonding configurations, structural versatility, and essential role in defining the mechanical, chemical, electronic, and quantum properties of materials. However, direct three-dimensional (3D) imaging of individual carbon atoms remains a long-standing challenge. Here, we use twisted bilayer graphene (TBG) as a model system and demonstrate ptychographic atomic electron tomography (pAET) for determining the 3D atomic coordinates of individual carbon atoms with a precision of 0.11 angstrom. The resulting 3D atomic model uncovers chiral lattice distortions driven by van der Waals interactions that exhibit meron-like and skyrmion-like structural textures. These findings provide direct insight into the interplay between 3D chiral lattice deformation and electronic properties in moire-engineered carbon systems. Beyond TBG, pAET offers a versatile approach for 3D atomic-scale imaging of carbon-based and other light-element materials that are central to advances in physics, chemistry, materials science, and nanotechnology.

cond-mat.mtrl-sci↗

arXiv2024

The Structure of Turbulence in Pulsatile Flow over Urban Canopies

The transport of energy, mass, and momentum in the atmospheric boundary layer (ABL) is regulated by coherent structures. Although past studies have primarily focused on stationary ABL flows, the majority of real-world ABL flows are non-stationary, and a thorough examination of coherent structures under such conditions is lacking. To fill this gap, this study examines the topological changes in ABL turbulence induced by non-stationarity and their effects on momentum transport. Results from a large-eddy simulation of pulsatile open channel flow over an array of surface-mounted cuboids are examined with a focus on the inertial sublayer, and contrasted to those from a corresponding constant pressure gradient case. The analysis reveals that flow pulsation primarily affects the ejection-sweep pattern. Inspection of the instantaneous turbulence structures, two-point autocorrelations, and conditionally-averaged flow fields shows that such a pattern is primarily influenced by the phase-dependent shear rate. From a turbulence structure perspective, this influence is attributed to the changes in the geometry of hairpin vortices. An increase (decrease) in the shear rate intensifies (relaxes) the hairpin vortices, leading to an increase (decrease) in the frequency of ejections and an amplification (reduction) of their percentage contribution to the total momentum flux. Moreover, the size of the hairpin vortex packets changes according to the hairpin vortices comprising them, while the packet inclination remains unaltered during the pulsatile cycle. Findings underscore the important impact of non-stationarity on the structure of ABL turbulence and associated mechanisms supporting momentum transport.

physics.flu-dyn↗

arXiv2023

Mean Flow and Turbulence in Unsteady Canopy Layers

Non-stationarity is the rule in the atmospheric boundary layer (ABL). Under such conditions, the flow may experience departures from equilibrium with the underlying surface stress, misalignment of shear stresses and strain rates, and three-dimensionality in turbulence statistics. Existing ABL flow theories are primarily established for statistically stationary flow conditions and cannot predict such behaviors. Motivated by this knowledge gap, this study analyzes the impact of time-varying pressure gradients on mean flow and turbulence over urban-like surfaces. A series of large-eddy simulations of pulsatile flow over cuboid arrays is performed, programmatically varying the oscillation amplitude $α$ and forcing frequency $ω$. The analysis focuses on both longtime-averaged and phase-dependent flow dynamics. Inspection of longtime-averaged velocity profiles reveals that the aerodynamic roughness length $z_0$ increases with $α$ and $ω$, whereas the displacement height $d$ appears to be insensitive to these parameters. In terms of phase-averaged flow statistics, it is found that $α$ primarily controls the oscillation amplitude of the streamwise velocity and Reynolds stresses, but has a negligible impact on their wall-normal structure. On the other hand, $ω$ determines the size of the region affected by the unsteady forcing, which identifies the so-called Stokes layer thickness $δ_s$. Within the Stokes layer, phase-averaged resolved Reynolds stress profiles feature substantial variations during the pulsatile cycle, and the turbulence is out of equilibrium with the mean flow. Two phenomenological models have been proposed that capture the influence of flow unsteadiness on $z_0$ and $δ_s$, respectively.

physics.flu-dyn↗

arXiv2021

Unsupervised feature selection via self-paced learning and low-redundant regularization

Much more attention has been paid to unsupervised feature selection nowadays due to the emergence of massive unlabeled data. The distribution of samples and the latent effect of training a learning method using samples in more effective order need to be considered so as to improve the robustness of the method. Self-paced learning is an effective method considering the training order of samples. In this study, an unsupervised feature selection is proposed by integrating the framework of self-paced learning and subspace learning. Moreover, the local manifold structure is preserved and the redundancy of features is constrained by two regularization terms. $L_{2,1/2}$-norm is applied to the projection matrix, which aims to retain discriminative features and further alleviate the effect of noise in the data. Then, an iterative method is presented to solve the optimization problem. The convergence of the method is proved theoretically and experimentally. The proposed method is compared with other state of the art algorithms on nine real-world datasets. The experimental results show that the proposed method can improve the performance of clustering methods and outperform other compared algorithms.

cs.LG↗

arXiv2021

SMOF: Squeezing More Out of Filters Yields Hardware-Friendly CNN Pruning

For many years, the family of convolutional neural networks (CNNs) has been a workhorse in deep learning. Recently, many novel CNN structures have been designed to address increasingly challenging tasks. To make them work efficiently on edge devices, researchers have proposed various structured network pruning strategies to reduce their memory and computational cost. However, most of them only focus on reducing the number of filter channels per layer without considering the redundancy within individual filter channels. In this work, we explore pruning from another dimension, the kernel size. We develop a CNN pruning framework called SMOF, which Squeezes More Out of Filters by reducing both kernel size and the number of filter channels. Notably, SMOF is friendly to standard hardware devices without any customized low-level implementations, and the pruning effort by kernel size reduction does not suffer from the fixed-size width constraint in SIMD units of general-purpose processors. The pruned networks can be deployed effortlessly with significant running time reduction. We also support these claims via extensive experiments on various CNN structures and general-purpose processors for mobile devices.

cs.CV↗