arXiv subjects

Chenyang Li

Publications and source records attributed to Chenyang Li.

At least 55 records · Page 3Linked to original sources

arXiv2025

Rethinking Occlusion in FER: A Semantic-Aware Perspective and Go Beyond

Facial expression recognition (FER) is a challenging task due to pervasive occlusion and dataset biases. Especially when facial information is partially occluded, existing FER models struggle to extract effective facial features, leading to inaccurate classifications. In response, we present ORSANet, which introduces the following three key contributions: First, we introduce auxiliary multi-modal semantic guidance to disambiguate facial occlusion and learn high-level semantic knowledge, which is two-fold: 1) we introduce semantic segmentation maps as dense semantics prior to generate semantics-enhanced facial representations; 2) we introduce facial landmarks as sparse geometric prior to mitigate intrinsic noises in FER, such as identity and gender biases. Second, to facilitate the effective incorporation of these two multi-modal priors, we customize a Multi-scale Cross-interaction Module (MCM) to adaptively fuse the landmark feature and semantics-enhanced representations within different scales. Third, we design a Dynamic Adversarial Repulsion Enhancement Loss (DARELoss) that dynamically adjusts the margins of ambiguous classes, further enhancing the model's ability to distinguish similar expressions. We further construct the first occlusion-oriented FER dataset to facilitate specialized robustness analysis on various real-world occlusion conditions, dubbed Occlu-FER. Extensive experiments on both public benchmarks and Occlu-FER demonstrate that our proposed ORSANet achieves SOTA recognition performance. Code is publicly available at https://github.com/Wenyuzhy/ORSANet-master.

cs.CV↗

arXiv2025

Fundamental Limits of Prompt Tuning Transformers: Universality, Capacity and Efficiency

We investigate the statistical and computational limits of prompt tuning for transformer-based foundation models. Our key contributions are prompt tuning on \emph{single-head} transformers with only a \emph{single} self-attention layer: (i) is universal, and (ii) supports efficient (even almost-linear time) algorithms under the Strong Exponential Time Hypothesis (SETH). Statistically, we prove that prompt tuning on such simplest possible transformers are universal approximators for sequence-to-sequence Lipschitz functions. In addition, we provide an exponential-in-$dL$ and -in-$(1/ε)$ lower bound on the required soft-prompt tokens for prompt tuning to memorize any dataset with 1-layer, 1-head transformers. Computationally, we identify a phase transition in the efficiency of prompt tuning, determined by the norm of the \emph{soft-prompt-induced} keys and queries, and provide an upper bound criterion. Beyond this criterion, no sub-quadratic (efficient) algorithm for prompt tuning exists under SETH. Within this criterion, we showcase our theory by proving the existence of almost-linear time prompt tuning inference algorithms. These fundamental limits provide important necessary conditions for designing expressive and efficient prompt tuning methods for practitioners.

cs.LG↗

arXiv2025

DAM-GT: Dual Positional Encoding-Based Attention Masking Graph Transformer for Node Classification

Neighborhood-aware tokenized graph Transformers have recently shown great potential for node classification tasks. Despite their effectiveness, our in-depth analysis of neighborhood tokens reveals two critical limitations in the existing paradigm. First, current neighborhood token generation methods fail to adequately capture attribute correlations within a neighborhood. Second, the conventional self-attention mechanism suffers from attention diversion when processing neighborhood tokens, where high-hop neighborhoods receive disproportionate focus, severely disrupting information interactions between the target node and its neighborhood tokens. To address these challenges, we propose DAM-GT, Dual positional encoding-based Attention Masking graph Transformer. DAM-GT introduces a novel dual positional encoding scheme that incorporates attribute-aware encoding via an attribute clustering strategy, effectively preserving node correlations in both topological and attribute spaces. In addition, DAM-GT formulates a new attention mechanism with a simple yet effective masking strategy to guide interactions between target nodes and their neighborhood tokens, overcoming the issue of attention diversion. Extensive experiments on various graphs with different homophily levels as well as different scales demonstrate that DAM-GT consistently outperforms state-of-the-art methods in node classification tasks.

cs.LG↗

arXiv2025

Unconditionally optimal error Estimate of a linearized Second-order Fully Discrete Finite Element Method for the bioconvection flows with concentration dependent viscosity

In this paper, the coupled and decoupled BDF2 finite element discrete schemes are obtained for the time-dependent bioconvection flows problem with concentration dependent viscosity, which consisting of the Navier-Stokes equation coupled with a linear convection-diffusion equation modeling the concentration of microorganisms in a culture fluid. The unconditionally optimal error estimate for the velocity and concentration in $L^2$-norm and $H^1$-norm are proved by using finite element approximations in space and finite differences in time. Finally, the numerical results for different viscosity are showed to support the theoretical analysis.

math.NA↗

arXiv2025

Error analysis of a Euler finite element scheme for Natural convection model with variable density

In this paper, we derive first-order Euler finite element discretization schemes for a time-dependent natural convection model with variable density (NCVD). The model is governed by the variable density Navier-Stokes equations coupled with a parabolic partial differential equation that describes the evolution of temperature. Stability and error estimate for the velocity, pressure, density and temperature in $L^2$-norm are proved by using finite element approximations in space and finite differences in time. Finally, the numerical results are showed to support the theoretical analysis.

math.NA↗

arXiv2025

Detection of Sleep Apnea-Hypopnea Events Using Millimeter-wave Radar and Pulse Oximeter

Obstructive Sleep Apnea-Hypopnea Syndrome (OSAHS) is a sleep-related breathing disorder associated with significant morbidity and mortality worldwide. The gold standard for OSAHS diagnosis, polysomnography (PSG), faces challenges in popularization due to its high cost and complexity. Recently, radar has shown potential in detecting sleep apnea-hypopnea events (SAE) with the advantages of low cost and non-contact monitoring. However, existing studies, especially those using deep learning, employ segment-based classification approach for SAE detection, making the task of event quantity estimation difficult. Additionally, radar-based SAE detection is susceptible to interference from body movements and the environment. Oxygen saturation (SpO2) can offer valuable information about OSAHS, but it also has certain limitations and cannot be used alone for diagnosis. In this study, we propose a method using millimeterwave radar and pulse oximeter to detect SAE, called ROSA. It fuses information from both sensors, and directly predicts the temporal localization of SAE. Experimental results demonstrate a high degree of consistency (ICC=0.9864) between AHI from ROSA and PSG. This study presents an effective method with lowload device for the diagnosis of OSAHS.

eess.SP↗

arXiv2025

Crystal nucleation and growth in high-entropy alloys revealed by atomic electron tomography

High-entropy alloys (HEAs) balance mixing entropy and intermetallic phase formation enthalpy, creating a vast compositional space for structural and functional materials (1-6). They exhibit exceptional strength-ductility trade-offs in metallurgy (4-10) and near-continuum adsorbate binding energies in catalysis (11-16). A deep understanding of crystal nucleation and growth in HEAs is essential for controlling their formation and optimizing their structural and functional properties. However, atomic-scale nucleation in HEAs challenges traditional theories based on one or two principal elements (17-23). The intricate interplay of structural and chemical orders among multiple principal elements further obscures our understanding of nucleation pathways (5,24-27). Due to the lack of direct three-dimensional (3D) atomic-scale observations, previous studies have relied on simulations and indirect measurements (28-32), leaving HEA nucleation and growth fundamentally elusive. Here, we advance atomic electron tomography (33,34) to resolve the 3D atomic structure and chemical composition of 7,662 HEA and 498 medium-entropy alloy nuclei at different nucleation stages. We observe local structural order that decreases from core to boundary, correlating with local chemical order. As nuclei grow, structural order improves. At later stages, most nuclei coalesce without misorientation, while some form coherent twin boundaries. To explain these experimental observations, we propose the gradient nucleation pathways model, in which the nucleation energy barrier progressively increases through multiple evolving intermediate states. We expect these findings to not only provide fundamental insights into crystal nucleation and growth in HEAs, but also offer a general framework for understanding nucleation mechanisms in other materials.

cond-mat.mtrl-sci↗

arXiv2025

Exploring Dynamic Transformer for Efficient Object Tracking

The speed-precision trade-off is a critical problem for visual object tracking which usually requires low latency and deployment on constrained resources. Existing solutions for efficient tracking mainly focus on adopting light-weight backbones or modules, which nevertheless come at the cost of a sacrifice in precision. In this paper, inspired by dynamic network routing, we propose DyTrack, a dynamic transformer framework for efficient tracking. Real-world tracking scenarios exhibit diverse levels of complexity. We argue that a simple network is sufficient for easy frames in video sequences, while more computation could be assigned to difficult ones. DyTrack automatically learns to configure proper reasoning routes for various inputs, gaining better utilization of the available computational budget. Thus, it can achieve higher performance with the same running speed. We formulate instance-specific tracking as a sequential decision problem and attach terminating branches to intermediate layers of the entire model. Especially, to fully utilize the computations, we introduce the feature recycling mechanism to reuse the outputs of predecessors. Furthermore, a target-aware self-distillation strategy is designed to enhance the discriminating capabilities of early predictions by effectively mimicking the representation pattern of the deep model. Extensive experiments on multiple benchmarks demonstrate that DyTrack achieves promising speed-precision trade-offs with only a single model. For instance, DyTrack obtains 64.9% AUC on LaSOT with a speed of 256 fps.

cs.CV↗

arXiv2025

SU-YOLO: Spiking Neural Network for Efficient Underwater Object Detection

Underwater object detection is critical for oceanic research and industrial safety inspections. However, the complex optical environment and the limited resources of underwater equipment pose significant challenges to achieving high accuracy and low power consumption. To address these issues, we propose Spiking Underwater YOLO (SU-YOLO), a Spiking Neural Network (SNN) model. Leveraging the lightweight and energy-efficient properties of SNNs, SU-YOLO incorporates a novel spike-based underwater image denoising method based solely on integer addition, which enhances the quality of feature maps with minimal computational overhead. In addition, we introduce Separated Batch Normalization (SeBN), a technique that normalizes feature maps independently across multiple time steps and is optimized for integration with residual structures to capture the temporal dynamics of SNNs more effectively. The redesigned spiking residual blocks integrate the Cross Stage Partial Network (CSPNet) with the YOLO architecture to mitigate spike degradation and enhance the model's feature extraction capabilities. Experimental results on URPC2019 underwater dataset demonstrate that SU-YOLO achieves mAP of 78.8% with 6.97M parameters and an energy consumption of 2.98 mJ, surpassing mainstream SNN models in both detection accuracy and computational efficiency. These results underscore the potential of SNNs for engineering applications. The code is available in https://github.com/lwxfight/snn-underwater.

cs.CV↗

arXiv2025

DeepOFormer: Deep Operator Learning with Domain-informed Features for Fatigue Life Prediction

Fatigue life characterizes the duration a material can function before failure under specific environmental conditions, and is traditionally assessed using stress-life (S-N) curves. While machine learning and deep learning offer promising results for fatigue life prediction, they face the overfitting challenge because of the small size of fatigue experimental data in specific materials. To address this challenge, we propose, DeepOFormer, by formulating S-N curve prediction as an operator learning problem. DeepOFormer improves the deep operator learning framework with a transformer-based encoder and a mean L2 relative error loss function. We also consider Stussi, Weibull, and Pascual and Meeker (PM) features as domain-informed features. These features are motivated by empirical fatigue models. To evaluate the performance of our DeepOFormer, we compare it with different deep learning models and XGBoost on a dataset with 54 S-N curves of aluminum alloys. With seven different aluminum alloys selected for testing, our DeepOFormer achieves an R2 of 0.9515, a mean absolute error of 0.2080, and a mean relative error of 0.5077, significantly outperforming state-of-the-art deep/machine learning methods including DeepONet, TabTransformer, and XGBoost, etc. The results highlight that our Deep0Former integrating with domain-informed features substantially improves prediction accuracy and generalization capabilities for fatigue life prediction in aluminum alloys.

cs.LG↗

arXiv2025

Stochastic resolution of identity to CC2 for large systems: Oscillator strength and ground state gradient calculations

An implementation of stochastic resolution of identity (sRI) approximation to CC2 oscillator strengths as well as ground state analytical gradients is presented. The essential 4-index electron repulsion integrals (ERIs) are contracted with a set of stochastic orbitals on the basis of the RI technique and the orbital energy differences in the denominators are decoupled with the Laplace transform. These lead to a significant scaling reduction from O(N^5) to O(N^3) for oscillator strengths and gradients with the size of the basis set, N. The gradients need a large number of stochastic orbitals with O(N^3), so we provide an additional O(N^4) version with better accuracy and smaller prefactor by adopting sRI partially. Such steep computational acceleration of nearly two or one order of magnitude is very attractive for large systems. This work is an extension to our previous implementations of sRI-CC2 ground and excited state energies and shows the feasibility of introducing sRI to CC2 properties beyond energies.

physics.chem-ph↗

arXiv2025

Fourier Circuits in Neural Networks and Transformers: A Case Study of Modular Arithmetic with Multiple Inputs

In the evolving landscape of machine learning, a pivotal challenge lies in deciphering the internal representations harnessed by neural networks and Transformers. Building on recent progress toward comprehending how networks execute distinct target functions, our study embarks on an exploration of the underlying reasons behind networks adopting specific computational strategies. We direct our focus to the complex algebraic learning task of modular addition involving $k$ inputs. Our research presents a thorough analytical characterization of the features learned by stylized one-hidden layer neural networks and one-layer Transformers in addressing this task. A cornerstone of our theoretical framework is the elucidation of how the principle of margin maximization shapes the features adopted by one-hidden layer neural networks. Let $p$ denote the modulus, $D_p$ denote the dataset of modular arithmetic with $k$ inputs and $m$ denote the network width. We demonstrate that a neuron count of $ m \geq 2^{2k-2} \cdot (p-1) $, these networks attain a maximum $ L_{2,k+1} $-margin on the dataset $ D_p $. Furthermore, we establish that each hidden-layer neuron aligns with a specific Fourier spectrum, integral to solving modular addition problems. By correlating our findings with the empirical observations of similar studies, we contribute to a deeper comprehension of the intrinsic computational mechanisms of neural networks. Furthermore, we observe similar computational mechanisms in attention matrices of one-layer Transformers. Our work stands as a significant stride in unraveling their operation complexities, particularly in the realm of complex algebraic tasks.

cs.LG↗

arXiv2025

Driven similarity renormalization group with a large active space: Applications to oligoacenes, zeaxanthin, and chromium dimer

We present a new implementation of the driven similarity renormalization group (DSRG) based on a density matrix renormalization group (DMRG) reference. The explicit build of high-order reduced density matrices is avoided by forming matrix-product-state compressed intermediates. This algorithm facilitates the application of DSRG second- and third-order perturbation theories to dodecacene with an active space of 50 electrons in 50 orbitals. This active space appears the largest employed to date within the framework of internally contracted multireference formalism. The DMRG-DSRG approach is applied to several challenging systems, including the singlet-triplet gaps ($Δ_{\rm ST}$) of oligoacenes ranging from naphthalene to dodecacene, the vertical excitation energies of zeaxanthin, and the ground-state potential energy curve (PEC) of Cr$_2$ molecule. Our best estimate for the vertical $Δ_{\rm ST}$ of dodecacene is 0.22 eV, showing an excellent agreement with that of the linearized adiabatic connection method (0.24 eV). For zeaxanthin, all DSRG schemes suggest the order of $\rm 2\, ^1 A_g^- < 1\, ^1 B_u^+ < 1\, ^1 B_u^-$ for excited states. Both the equilibrium and the shoulder regions of the Cr$_2$ PEC are reasonably reproduced by the linearized DSRG with one- and two-body operators.

physics.chem-ph↗

arXiv2025

MetaDesigner: Advancing Artistic Typography Through AI-Driven, User-Centric, and Multilingual WordArt Synthesis

MetaDesigner introduces a transformative framework for artistic typography synthesis, powered by Large Language Models (LLMs) and grounded in a user-centric design paradigm. Its foundation is a multi-agent system comprising the Pipeline, Glyph, and Texture agents, which collectively orchestrate the creation of customizable WordArt, ranging from semantic enhancements to intricate textural elements. A central feedback mechanism leverages insights from both multimodal models and user evaluations, enabling iterative refinement of design parameters. Through this iterative process, MetaDesigner dynamically adjusts hyperparameters to align with user-defined stylistic and thematic preferences, consistently delivering WordArt that excels in visual quality and contextual resonance. Empirical evaluations underscore the system's versatility and effectiveness across diverse WordArt applications, yielding outputs that are both aesthetically compelling and context-sensitive.

cs.AI↗

arXiv2025

When Can We Solve the Weighted Low Rank Approximation Problem in Truly Subquadratic Time?

The weighted low-rank approximation problem is a fundamental numerical linear algebra problem and has many applications in machine learning. Given a $n \times n$ weight matrix $W$ and a $n \times n$ matrix $A$, the goal is to find two low-rank matrices $U, V \in \mathbb{R}^{n \times k}$ such that the cost of $\| W \circ (U V^\top - A) \|_F^2$ is minimized. Previous work has to pay $Ω(n^2)$ time when matrices $A$ and $W$ are dense, e.g., having $Ω(n^2)$ non-zero entries. In this work, we show that there is a certain regime, even if $A$ and $W$ are dense, we can still hope to solve the weighted low-rank approximation problem in almost linear $n^{1+o(1)}$ time.

cs.CC↗

arXiv2025

VibNet: Vibration-Boosted Needle Detection in Ultrasound Images

Precise percutaneous needle detection is crucial for ultrasound (US)-guided interventions. However, inherent limitations such as speckles, needle-like artifacts, and low resolution make it challenging to robustly detect needles, especially when their visibility is reduced or imperceptible. To address this challenge, we propose VibNet, a learning-based framework designed to enhance the robustness and accuracy of needle detection in US images by leveraging periodic vibration applied externally to the needle shafts. VibNet integrates neural Short-Time Fourier Transform and Hough Transform modules to achieve successive sub-goals, including motion feature extraction in the spatiotemporal space, frequency feature aggregation, and needle detection in the Hough space. Due to the periodic subtle vibration, the features are more robust in the frequency domain than in the image intensity domain, making VibNet more effective than traditional intensity-based methods. To demonstrate the effectiveness of VibNet, we conducted experiments on distinct \textit{ex vivo} porcine and bovine tissue samples. The results obtained on porcine samples demonstrate that VibNet effectively detects needles even when their visibility is severely reduced, with a tip error of $1.61\pm1.56~mm$ compared to $8.15\pm9.98~mm$ for UNet and $6.63\pm7.58~mm$ for WNet, and a needle direction error of $1.64\pm1.86^{\circ}$ compared to $9.29\pm15.30^{\circ}$ for UNet and $8.54\pm17.92^{\circ}$ for WNet. Code: https://github.com/marslicy/VibNet.

eess.IV↗

arXiv2025

Rethinking Tokenized Graph Transformers for Node Classification

Node tokenized graph Transformers (GTs) have shown promising performance in node classification. The generation of token sequences is the key module in existing tokenized GTs which transforms the input graph into token sequences, facilitating the node representation learning via Transformer. In this paper, we observe that the generations of token sequences in existing GTs only focus on the first-order neighbors on the constructed similarity graphs, which leads to the limited usage of nodes to generate diverse token sequences, further restricting the potential of tokenized GTs for node classification. To this end, we propose a new method termed SwapGT. SwapGT first introduces a novel token swapping operation based on the characteristics of token sequences that fully leverages the semantic relevance of nodes to generate more informative token sequences. Then, SwapGT leverages a Transformer-based backbone to learn node representations from the generated token sequences. Moreover, SwapGT develops a center alignment loss to constrain the representation learning from multiple token sequences, further enhancing the model performance. Extensive empirical results on various datasets showcase the superiority of SwapGT for node classification.

cs.LG↗

arXiv2025

Fusion of Millimeter-wave Radar and Pulse Oximeter Data for Low-burden Diagnosis of Obstructive Sleep Apnea-Hypopnea Syndrome

Objective: The aim of the study is to develop a novel method for improved diagnosis of obstructive sleep apnea-hypopnea syndrome (OSAHS) in clinical or home settings, with the focus on achieving diagnostic performance comparable to the gold-standard polysomnography (PSG) with significantly reduced monitoring burden. Methods: We propose a method using millimeter-wave radar and pulse oximeter for OSAHS diagnosis (ROSA). It contains a sleep apnea-hypopnea events (SAE) detection network, which directly predicts the temporal localization of SAE, and a sleep staging network, which predicts the sleep stages throughout the night, based on radar signals. It also fuses oxygen saturation (SpO2) information from the pulse oximeter to adjust the score of SAE detected by radar. Results: Experimental results on a real-world dataset (>800 hours of overnight recordings, 100 subjects) demonstrated high agreement (ICC=0.9870) on apnea-hypopnea index (AHI) between ROSA and PSG. ROSA also exhibited excellent diagnostic performance, exceeding 90% in accuracy across AHI diagnostic thresholds of 5, 15 and 30 events/h. Conclusion: ROSA improves diagnostic accuracy by fusing millimeter-wave radar and pulse oximeter data. It provides a reliable and low-burden solution for OSAHS diagnosis. Significance: ROSA addresses the limitations of high complexity and monitoring burden associated with traditional PSG. The high accuracy and low burden of ROSA show its potential to improve the accessibility of OSAHS diagnosis among population.

eess.SP↗