arXiv subjects

Xiangyang Wang

Publications and source records attributed to Xiangyang Wang.

18 recordsLinked to original sources

arXiv2026

Audio-Anchored Fusion of Multi-Ratio DiT Reconstruction Residuals for Cross-Domain Audio Deepfake Detection

Audio deepfake detectors often degrade when generators, corpora, or recording conditions change. We use a Diffusion Transformer (DiT), trained only on bona fide speech, as a frozen reconstruction probe. Reconstructions at masking ratios 0.5, 0.75, and 0.9 yield explicit multi-ratio residual maps. Because these residuals are domain sensitive, our audio-anchored detector passes the projected frozen-WavLM auditory representation into the fusion sum without gate-based attenuation and uses residuals only as a scalar-gated additive correction. The pre-specified seed-42 run obtains 6.5442% EER / 0.18456 min-DCF on ASVspoof 5 Eval and 13.8372% / 0.36921 on ITW Full; three-seed means are 6.8885 (0.3308)% and 15.3328 (2.0719)%. The latter is below a separately optimized WavLM-ResNet18 reference under both supervision settings. Auxiliary supervision raises dynamic competitive fusion from 18.4007% to 25.2968% mean ITW EER, worsening all three seeds. The results support reconstruction residuals as complementary evidence and motivate a non-competitive auditory path for ASVspoof 5-to-ITW transfer, without claiming a componentwise causal ablation of anchoring alone.

cs.SD↗

arXiv2026

Hybrid Pinching-Fluid Antenna Assisted Wireless Communications: Modeling and Performance Analysis

Reconfigurable-antenna systems have received increasing attention for their ability to adapt wireless channels. However, existing architectures exhibit scenario-dependent limitations: fluid antennas provide strong diversity gains in rich-scattering environments but offer limited benefits under line-of-sight (LoS)-dominant conditions, while pinching antennas can effectively reduce path loss by adjusting the radiation point along a waveguide, yet perform poorly in severe non-LoS (NLoS) scenarios. This letter proposes a hybrid pinching-fluid antenna system (HPFAS), where pinching antenna (PA) is employed at the transmitter and a fluid antenna (FA) is used at the receiver to jointly exploit LoS enhancement and spatial diversity. A tractable channel model is developed, and outage probability expressions are derived for both single-user and multi-user scenarios. Simulation results validate the analysis and show that the proposed HPFAS consistently outperforms systems using only pinching antennas or only fluid antennas under various propagation conditions.

eess.SP↗

arXiv2026

Every connected subcubic graph except the Petersen graph is packing $(1,1,2,2)$-colorable

For a non-decreasing sequence $S = (s_1, s_2, \ldots, s_k)$ of positive integers, a packing $S$-coloring of a graph $G$ is a partition of $V(G)$ into $V_1, V_2, \ldots, V_k$ such that each $V_i$ has pairwise distance at least $s_i+1$. The packing chromatic number (PCN) of a graph $G$ is the minimum $k$ such that $G$ has a packing $(1,2, \ldots, k)$-coloring. The $1$-subdivision of $G$ is obtained by replacing each edge of $G$ with a path of two edges. In 2016, Gastineau and Togni asked an open question whether the $1$-subdivision of every subcubic graph has PCN at most $5$, and later Bre\v sar, Klav\v zar, Rall, and Wash conjectured it is true. Balogh, Kostochka, and Liu proved the first upper bound of $8$, and it was later improved to $6$ by Liu, Zhang, and Zhang. In this paper, we prove that every connected subcubic graph except the Petersen graph is packing $(1,1,2,2)$-colorable. Our result implies a solution to the conjecture of Bre\v sar, Klav\v zar, Rall, and Wash, and answers the question of Gastineau and Togni in the affirmative. Furthermore, our result answers an open question of Kostochka and Liu and solves a conjecture of Liu, Zhang, and Zhang.

math.CO↗

arXiv2026

Monochromatic triangle-tilings in dense graphs without large independent sets

Given two graphs $H$ and $G$, an $H$-tiling is a family of vertex-disjoint copies of $H$ in $G$. A perfect $H$-tiling covers all vertices of $G$. The Corradi-Hajnal theorem (1963) states that an $n$-vertex graph $G$ with minimum degree $\delta(G)\ge 2n/3$ contains a perfect triangle-tiling. For an $n$-vertex graph $G$ with independence number $\alpha(G)=o(n)$, Balogh, Molla and Sharifzadeh (Random Structures & Algorithms, 2016) showed that a minimum degree of $(\frac12+o(1))n$ forces a perfect triangle-tiling. In a 2-edge-colored graph, Balogh, Freschi, Treglown (European J. Combin. 2026) determined the (asymptotic) minimum degree threshold for forcing a strong or weak monochromatic triangle-tiling covering a prescribed proportion of the vertices: a strong tiling requires all triangles to be in the same color class, while a weak tiling only requires each triangle to be monochromatic. In this paper, we combine the conditions from these two lines of work and prove that every $2$-edge-colored $n$-vertex graph $G$ with $\alpha(G)=o(n)$ contains a weak monochromatic triangle-tiling $\Gamma$ of size \[ |\Gamma|\ge \begin{cases} 2\delta(G)-n-o(n), & \text{if }\frac12 n\le \delta(G)\le \frac35 n,\\[2mm] \delta(G)/3-o(n), & \text{if }\delta(G)>\frac35 n. \end{cases} \] Both bounds are asymptotically optimal. We use the degree form regularity lemma in our proof.

math.CO↗

arXiv2025

A new multiscale modeling approach to unravel the influence of interlayer sp3 bonds on the nonlinear large-deformation and fracture behaviors of 2D carbon nanostructures under tension

To delve deeply into the nonlinear large-deformation and fracture behaviors of 2D carbon nanostructures (2D CNs), including bilayer graphene, diamane, and their transitional structures, this paper introduces a multiscale auxiliary nodes (MAN) method rooted in atomic structures and potentials. This approach simulates 2D CNs by constructing two virtual continuum sheets with high-order continuity. The moving least squares (MLS) approximation is employed to facilitate the transformation between atomic displacements and nodal displacements, thereby converting atomic potential energy into strain energy within the continuum model. Through iterative solutions of nonlinear stiffness equations, the equilibrium configuration of the system under specified loading conditions can be obtained. The flexibility in the density and arrangement of nodes allows for a smooth and seamless cross-scale transition from discrete atomic structures to a continuum model. Numerical simulations demonstrate that MAN method accurately predicts the nonlinear large-deformation and fracture behaviors of 2D CNs. The Young's modulus and shear modulus of diamane in both zigzag and armchair directions closely approach those of diamond and are notably higher than those of graphene. Furthermore, the quantity and distribution of interlayer sp3 bonds significantly influence the fracture behavior of 2D CNs, with strategic placement of these bonds effectively enhancing the tensile strength of the structures.

cond-mat.mtrl-sci↗

arXiv2025

WisPaper: Your AI Scholar Search Engine

We present \textsc{WisPaper}, an end-to-end agent system that transforms how researchers discover, organize, and track academic literature. The system addresses two fundamental challenges. (1)~\textit{Semantic search limitations}: existing academic search engines match keywords but cannot verify whether papers truly address complex research questions; and (2)~\textit{Workflow fragmentation}: researchers must manually stitch together separate tools for discovery, organization, and monitoring. \textsc{WisPaper} tackles these through three integrated modules. \textbf{Scholar Search} combines rapid keyword retrieval with \textit{Deep Search}, in which an agentic model, \textsc{WisModel}, validates candidate papers against user queries through structured reasoning. Discovered papers flow seamlessly into \textbf{Library} with one click, where systematic organization progressively builds a user profile that sharpens the recommendations of \textbf{AI Feeds}, which continuously surfaces relevant new publications and in turn guides subsequent exploration, closing the loop from discovery to long-term awareness. On TaxoBench, \textsc{WisPaper} achieves 22.26\% recall, surpassing the O3 baseline (20.92\%). Furthermore, \textsc{WisModel} attains 93.70\% validation accuracy, effectively mitigating retrieval hallucinations.

cs.IR↗

arXiv2025

Divisible subdivisions of graphs in subdivisions of complete graphs

Let $\mathbb{Z}_q$ denote the cyclic group of order $q$. A $\mathbb{Z}_q$-edge-weighted $K_f$ is the complete graph $K_f$ equipped with a weight function $\omega : E(K_f) \to \mathbb{Z}_q$. A subdivision of a graph $H$ in a $\mathbb{Z}_q$-edge-weighted $K_f$ is called a $q$-divisible subdivision of $H$ if every subdivision path has weight congruent to zero modulo $q$. Let $q\ge 2$ be an integer and let $H$ be a graph with $n$ vertices and $m$ edges. Define $s_q(H)$ to be the smallest number $f$ such that every $\mathbb{Z}_q$-edge-weighted $K_{f}$ contains a $q$-divisible subdivision of $H$. Das, Dragani\'c, and Steiner raised the following question (Problem 4.1 in [Tight bounds for divisible subdivisions, J. Combin. Theory, Ser. B 165 (2024) 1-19]): Given $q\in\mathbb{N}$ and a subcubic graph $H$ with $n$ vertices and $m$ edges, is it true $s_q(H)= m(q - 1) + n$? They also established the upper bound $s_q(H)\le 7mq+8n+14q$ for such a graph $H$. In this paper, we improve this bound by showing that $s_q(H)\le (2q - 1)m + 2n - 1 + 4q$, and establishing a sharper bound $s_p(H)\le \frac{3p - 1}{2}m - \frac{p - 1}{2}n + \frac{p + 1}{2}$ for prime $p$ and connected $H$. We resolve this problem in the case $q=2$ by proving that $s_2(H) = m + n$ for any 5-degenerate graph $H$, and in the case $q\ge 2$ and $T$ being a tree, by showing that $s_q(T) = nq - q + 1$. Let $s_q(H,t)$ be the minimum number $f$ such that every $\mathbb{Z}_q$-edge-weighted $K_f$ contains a $q$-divisible $t$-subdivision of $H$, where a $t$-subdivision of $H$ is a subdivision of $H$ such that each edge of $H$ is subdivided exactly $t$ times. We also prove that $s_2(H,1)= m + n$, where $H$ is a tree or a cycle on $n$ vertices with $m$ edges.

math.CO↗

arXiv2025

MOAT: Evaluating LMMs for Capability Integration and Instruction Grounding

Large multimodal models (LMMs) have demonstrated significant potential as generalists in vision-language (VL) tasks. However, adoption of LMMs in real-world tasks is hindered by their poor performance in tasks that require a combination of VL capabilities, as well as in tasks that involve the grounding of complex text or visual instructions. To thoroughly investigate this gap and its underlying causes, we propose MOAT, a diverse benchmark with 1005 complex real-world vision questions that are straightforward for humans but challenging for LMMs. Specifically, the tasks in MOAT require LMMs to engage in generalist problem solving by integrating VL capabilities such as reading text, counting, understanding spatial relations, grounding textual and visual instructions, etc. All these abilities fit into a taxonomy proposed by us that contains 9 VL capabilities, enabling MOAT to provide a fine-grained view of LMMs' strengths and weaknesses. Besides, MOAT is the first benchmark to explicitly evaluate LMMs' ability to ground complex text and visual instructions, which is essential for many real-world applications. We evaluated 17 proprietary and open source LMMs, finding that the best performing LMM (Gemini 2.5 Pro) achieved only 44% accuracy, far below what would be acceptable in real-world applications. To guide future model development, we analyze common trends in our results and discuss the underlying causes of poor performance, focusing on the impact of text-centric reasoning, which VL capabilities form bottlenecks in complex tasks, and the potential harmful effects of tiling. Code and data are available at https://cambrian-yzt.github.io/MOAT/.

cs.CL↗

arXiv2025

Computing with Smart Rings: A Systematic Literature Review

A smart ring is a wearable electronic device in the form of a ring that incorporates diverse sensors and computing technologies to perform a variety of functions. Designed for use with fingers, smart rings are capable of sensing more subtle and abundant hand movements, thus making them a good platform for interaction. Meanwhile, fingers are abundant with blood vessels and nerve endings and accustomed to wearing rings, providing an ideal site for continuous health monitoring through smart rings, which combine comfort with the ability to capture vital biometric data, making them suitable for all-day wear. We collected in total of 206 smart ring-related publications and conducted a systematic literature review. We provide a taxonomy regarding the sensing and feedback modalities, applications, and phenomena. We review and categorize these literatures into four main areas: (1) interaction - input, (2) interaction - output, (3) passive sensing - in body feature, (4) passive sensing - out body activity. This comprehensive review highlights the current advancements within the field of smart ring and identifies potential areas for future research.

cs.HC↗

arXiv2021

Convolutional Neural Network Pruning with Structural Redundancy Reduction

Convolutional neural network (CNN) pruning has become one of the most successful network compression approaches in recent years. Existing works on network pruning usually focus on removing the least important filters in the network to achieve compact architectures. In this study, we claim that identifying structural redundancy plays a more essential role than finding unimportant filters, theoretically and empirically. We first statistically model the network pruning problem in a redundancy reduction perspective and find that pruning in the layer(s) with the most structural redundancy outperforms pruning the least important filters across all layers. Based on this finding, we then propose a network pruning approach that identifies structural redundancy of a CNN and prunes filters in the selected layer(s) with the most redundancy. Experiments on various benchmark network architectures and datasets show that our proposed approach significantly outperforms the previous state-of-the-art.

cs.CV↗

arXiv2019

Investigating Channel Pruning through Structural Redundancy Reduction -- A Statistical Study

Most existing channel pruning methods formulate the pruning task from a perspective of inefficiency reduction which iteratively rank and remove the least important filters, or find the set of filters that minimizes some reconstruction errors after pruning. In this work, we investigate the channel pruning from a new perspective with statistical modeling. We hypothesize that the number of filters at a certain layer reflects the level of 'redundancy' in that layer and thus formulate the pruning problem from the aspect of redundancy reduction. Based on both theoretic analysis and empirical studies, we make an important discovery: randomly pruning filters from layers of high redundancy outperforms pruning the least important filters across all layers based on the state-of-the-art ranking criterion. These results advance our understanding of pruning and further testify to the recent findings that the structure of the pruned model plays a key role in the network efficiency as compared to inherited weights.

cs.CV↗

arXiv2019

Single-shot Channel Pruning Based on Alternating Direction Method of Multipliers

Channel pruning has been identified as an effective approach to constructing efficient network structures. Its typical pipeline requires iterative pruning and fine-tuning. In this work, we propose a novel single-shot channel pruning approach based on alternating direction methods of multipliers (ADMM), which can eliminate the need for complex iterative pruning and fine-tuning procedure and achieve a target compression ratio with only one run of pruning and fine-tuning. To the best of our knowledge, this is the first study of single-shot channel pruning. The proposed method introduces filter-level sparsity during training and can achieve competitive performance with a simple heuristic pruning criterion (L1-norm). Extensive evaluations have been conducted with various widely-used benchmark architectures and image datasets for object classification purpose. The experimental results on classification accuracy show that the proposed method can outperform state-of-the-art network pruning works under various scenarios.

cs.CV↗

arXiv2019

Speeding up convolutional networks pruning with coarse ranking

Channel-based pruning has achieved significant successes in accelerating deep convolutional neural network, whose pipeline is an iterative three-step procedure: ranking, pruning and fine-tuning. However, this iterative procedure is computationally expensive. In this study, we present a novel computationally efficient channel pruning approach based on the coarse ranking that utilizes the intermediate results during fine-tuning to rank the importance of filters, built upon state-of-the-art works with data-driven ranking criteria. The goal of this work is not to propose a single improved approach built upon a specific channel pruning method, but to introduce a new general framework that works for a series of channel pruning methods. Various benchmark image datasets (CIFAR-10, ImageNet, Birds-200, and Flowers-102) and network architectures (AlexNet and VGG-16) are utilized to evaluate the proposed approach for object classification purpose. Experimental results show that the proposed method can achieve almost identical performance with the corresponding state-of-the-art works (baseline) while our ranking time is negligibly short. In specific, with the proposed method, 75% and 54% of the total computation time for the whole pruning procedure can be reduced for AlexNet on CIFAR-10, and for VGG-16 on ImageNet, respectively. Our approach would significantly facilitate pruning practice, especially on resource-constrained platforms.

cs.CV↗

arXiv2018

Light rays and waves on geodesic lenses

Starting from well-known absolute instruments for perfect imaging, we introduce a type of rotational-symmetrical compact closed manifolds, namely geodesic lenses. We demonstrate that light rays confined on geodesic lenses are closed trajectories. While for optical waves, the spectrum of geodesic lens is (at least approximately) degenerate and equidistant with numerical methods. Based on this property, we show a periodical evolution of optical waves and quantum waves on geodesic lenses. Moreover, we fabricate two geodesic lenses in sub-micrometer scale, where curved light rays are observed with high accurate precision. Our results may offer a new platform to investigate light propagation on curved surfaces.

physics.optics↗

arXiv2018

Reversible Wavefront Shaping Between Gaussian and Airy Beams By Mimicking Gravitational Field

In this paper, we experimentally demonstrate reversible wavefront shaping through mimicking gravitational field. A gradient-index micro-structured optical waveguide with special refractive index profile was constructed whose effective index satisfying a gravitational field profile. Inside the waveguide, an incident broad Gaussian beam is firstly transformed into an accelerating beam, and the generated accelerating beam is gradually changed back to a Gaussian beam afterwards. To validate our experiment, we performed full-wave continuum simulations that agree with the experimental results. Furthermore, a theoretical model was established to describe the evolution of the laser beam based on Landau's method, showing that the accelerating beam behaves like the Airy beam in the small range in which the linear potential approaches zero. To our knowledge, such a reversible wavefront shaping technique has not been reported before.

physics.optics↗

arXiv2017

Self-focusing and the Talbot effect in conformal transformation optics

Transformation optics (TO) has been used to propose various novel optical devices. With the help of metamaterials, several intriguing designs, such as invisibility cloaks, have been implemented. However, as the basic units should be much smaller than the working wavelengths to achieve the effective material parameters, and the sizes of devices should be much larger than the wavelengths of illumination to work within the light-ray approximation, it is a big challenge to implement an experimental system that works simultaneously for both geometric optics and wave optics. In this letter, by using a gradient-index micro-structured optical waveguide, we realize a device of conformal transformation optics (CTO) and demonstrate its self-focusing property for geometry optics and Talbot effect for wave optics. In addition, the Talbot effect in such a system has a potential application to transfer digital information without diffraction. Our findings demonstrate the photon controlling ability of CTO in a feasible experiment system.

physics.optics↗

arXiv2011

Meshless simulation for thermo-mechanical properties of single-walled carbon nanotubes based on the thermal-related higher order Cauchy-Born rule

In the present paper, a temperature-dependent meshless numerical framework based on the thermo-related quasi-continuum constitutive model is developed for predicting the thermal mechanical properties of single-walled carbon nanotubes (SWCNTs) at finite temperature. The extended thermal-related higher order Cauchy-Born (THCB) rule included second order deformation gradient relates the deformation of bond vectors of the atomic system and that of the continuous medium, which can capture the curvature effect of carbon nanotubes (CNTs) conveniently. Helmholtz free energy is employed to allow for the thermal effect of SWCNTs. In the meshless numerical implementations of the theory, the Newton iteration method is applied to find the equilibrium configuration of a SWCNT subjected to large deformation at a prescribed temperature only with the nodal displace parameters as optimization variables. The finite deformation behaviors of armchair and zigzag SWCNTs under axial compression and torsion are tested. It is shown that the simulation results are in good agreement with those obtained by molecular dynamic methods even with fewer meshless nodes used.

cond-mat.mtrl-sci↗

arXiv2011

Kato's inequality and Liouville theorems on locally finite graphs

In this paper we study the Kato' inequality on locally finite graph. We also study the application of Kato inequality to Ginzburg-Landau equations on such graphs. Interesting properties of Schrodinger equation and a Liouville type theorem are also derived.

math.AP↗