Search arXiv⌕ Search

arXiv subjects

Hong Zhou

Publications and source records attributed to Hong Zhou.

At least 19 recordsLinked to original sources
arXiv2026

Sonicmesh: Enhancing 3D Human Mesh Reconstruction in Vision-Impaired Environments With Acoustic Signals

3D human mesh reconstruction (HMR) from RGB images often degrades under poor illumination, occlusion, and non-line-of-sight conditions. Acoustic sensing provides complementary spatial cues but suffers from low spatial resolution. We propose SonicMesh, which, to the best of our knowledge, is the first acoustic--visual framework for robust 3D human mesh reconstruction. SonicMesh first converts ultrasonic echoes into range--azimuth acoustic images through an Inverse Synthetic Aperture Radar (ISAR)-based imaging process. It then introduces a cross-dimensional anatomical registration module that maps modality-specific 2D joint features into a common canonical 3D human space. The registered anatomical representations are further integrated with acoustic and visual features through a two-stage fusion network for final mesh reconstruction. Experiments demonstrate that SonicMesh achieves accurate and robust 3D human reconstruction across normal, poor-light, occluded, and non-line-of-sight environments, consistently outperforming existing RGB-, radio-frequency (RF)-, and mmWave-based approaches under challenging sensing conditions.

cs.CV↗
arXiv2026

Improved Bounds for the Bilu--Linial Conjecture via Spectral Recovery from Mixed Determinantal Polynomials

The Bilu--Linial conjecture asks whether every finite $d$-regular graph with $d \geq 2$ admits an edge signing $σ$ whose signed adjacency matrix $A_σ$ has spectral radius at most $2\sqrt{d-1}$. We prove that every signing meeting the mixed-root condition $r_{A_σ}\leq\sqrt{2(d-1)}$ satisfies \[ ρ(A_σ) < \frac{3+\sqrt5}{2}\sqrt{d-1}, \] where $r_{A_σ}$ is the largest root of the mixed determinantal polynomial $χ[A_σ,-A_σ]$. The interlacing theorem of Ravichandran and Srivastava guarantees a signing satisfying the mixed-root condition, so our result improves the coefficient $2\sqrt2$ in their two-sided spectral bound. In the proof, we construct a positive matrix-valued probability measure supported on the roots of $χ[A_σ,-A_σ]$. The second moment gives a simple matrix inequality $A_σ^2 + dI \preceq 4r_{A_σ}^2I$, which yields a preliminary coefficient $\sqrt{7}$. Estimates for the fourth moment use information about short walks to obtain the coefficient $(3+\sqrt{5})/2$. With more graph structural assumptions, the coefficient improves to $\sqrt6$ for triangle-free graphs and to $\sqrt{(5+3\sqrt5)/2}$ for graphs of girth at least five. As a result of independent interest, we extend the construction to $χ[A_1,\ldots,A_k]$ for Hermitian matrices $A_1,\ldots,A_k$ with zero diagonal, and compute the first two moments explicitly. Finally, an explicit signing of $K_8$ shows that the mixed-root condition alone cannot guarantee a coefficient below $(4+\sqrt5)/\sqrt6$.

math.CO↗
arXiv2026

Derandomizing Matrix Concentration Inequalities from Free Probability

Recently, sharp matrix concentration inequalities~\cite{BBvH23,BvH24} were developed using the theory of free probability. In this work, we design polynomial time deterministic algorithms to construct outcomes that satisfy the guarantees of these inequalities. As direct consequences, we obtain polynomial time deterministic algorithms for the matrix Spencer problem~\cite{BJM23} and for constructing near-Ramanujan graphs. Our proofs show that the concepts and techniques in free probability are useful not only for mathematical analyses but also for efficient computations.

cs.DS↗
arXiv2026

Are Android GUI Agents Robust Against Runtime Anomalies? AnTrap: Evaluating Agents in Dynamic Adversarial Environments

GUI agents often encounter dynamic anomalies when deployed on Android devices, from unexpected pop-ups to action misuse, yet existing benchmarks lack systematic evaluation of agent robustness against runtime anomalies. We introduce AnTrap, a comprehensive benchmark that injects dynamic perturbations into agent execution trajectories. We propose a taxonomy organizing real-world anomalies into four layers (State, Thinking, Action and Round) with ten fine-grained subcategories, and develop a construction pipeline that preserves task solvability while introducing realistic adversarial conditions. Evaluating 16 leading GUI models, we reveal universal vulnerability to dynamic anomalies, with even the strongest models suffering significant performance degradation. Furthermore, we conduct GRPO training in both original and adversarial environments to validate our benchmark, separating environment-learnable anomalies from reasoning-bottlenecked ones. Our findings show that while single-step traps at state and action layers are largely addressable through adversarial reinforcement learning, deep contextual traps, like state deadlock, expose intrinsic limitations that cannot be resolved by training in environments with traps alone.

cs.AI↗
arXiv2026

Accurate Computation of Activated Volume in Electromagnetic Heating

In electromagnetic heating and other applications, we need to compute the volume enclosed by an isosurface of the 3D temperature distribution that is numerically represented on a rectangular grid. This situation arises naturally when the temperature distribution is obtained by solving a partial differential equation numerically using a finite difference method. Given the temperature distribution on the 3D grid, the isosurface of a prescribed value is represented approximately by a triangulation, a collection of triangles with vertices on the grid lines. The vertices are determined by a linear interpolation to approximate the locations where the temperature is at the prescribed level. The region enclosed by the isosurface is approximated by that enclosed by the triangulation, which is a set of tetrahedrons. The enclosed volume is approximated by summing those of tetrahedrons. This volume approximation is analogous to approximating a curve using line segments and is limited in accuracy. In this study, we combine extrapolation with the triangulation approximation to develop a more accurate method for computing the volume enclosed by an isosurface of the 3D temperature distribution on a rectangular grid.

math-ph↗
arXiv2026

Android Coach: Improve Online Agentic Training Efficiency with Single State Multiple Actions

Online reinforcement learning (RL) serves as an effective method for enhancing the capabilities of Android agents. However, guiding agents to learn through online interaction is prohibitively expensive due to the high latency of emulators and the sample inefficiency of existing RL algorithms. We identify a fundamental limitation in current approaches: the Single State Single Action paradigm, which updates the policy with one-to-one state-action pairs from online one-way rollouts without fully exploring each costly emulator state. In this paper, we propose Android Coach, a novel framework that shifts the training paradigm to Single State Multiple Actions, allowing the agent to sample and utilize multiple actions for a single online state. We enable this without additional emulator overhead by learning a critic that estimates action values. To ensure the critic serves as a reliable coach, we integrate a process reward model and introduce a group-wise advantage estimator based on the averaged critic outputs. Extensive experiments demonstrate the effectiveness and efficiency of Android Coach: it achieves 7.5% and 8.3% success rate improvements on AndroidLab and AndroidWorld over UI-TARS-1.5-7B, and attains 1.4x higher training efficiency than Single State Single Action methods PPO and GRPO at matched success rates.

cs.LG↗
arXiv2026

An Analytical Framework for Frequency-Dependent Electromagnetic Power Absorption in Biological Tissues

As exposure to electromagnetic waves becomes increasingly widespread, it is important to quantify how incident fields couple into biological tissue and where absorbed energy is deposited. This work presents an analytical, physics based framework derived from Maxwell's equations to model the propagation of a normally incident electromagnetic plane wave within homogeneous, lossy dielectric biological tissues. Closed-form expressions for the electric and magnetic fields are derived, enabling the determination of frequency-dependent power reflectance and transmittance at the air-tissue interface, as well as the power absorption coefficient and penetration depth within the medium. Using complex relative permittivity data from the literature, we examine six tissue types across a broad frequency range (1 MHz to 100 GHz). The results demonstrate that higher water content significantly increases dielectric loss and reduces penetration depth. Conversely, low-water tissues (e.g., non-infiltrated fat) exhibit lower attenuation and deeper penetration. Frequency is shown to be a dominant driver of this behavior, with higher frequencies shifting the power budget from reflection-limited coupling toward highly superficial absorption. These findings provide a foundation basis for exposure assessments and the design of emerging electromagnetic technologies.

physics.bio-ph↗
arXiv2026

Asymptotic Effects of Incident Angle and Lateral Conduction in Electromagnetic Skin Heating

Previously we derived the leading term asymptotic solution of temperature distribution in skin heating by an electromagnetic beam at an arbitrary incident angle. The asymptotic analysis is based on that the penetration depth of the beam into skin is much smaller than the size of beam cross-section. It allows arbitrary incident angle. We expand the temperature in powers of the small depth to lateral scale ratio. The incident angle affects all terms in the expansion while the lateral heat conduction appears only in terms of positive even powers. The previously obtained leading term solution captures only the main effect of incident angle. The main effect of lateral heat conduction is contained in the second order term, which is mathematically negligible in the limit of small depth to lateral scale ratio. At a moderate length scale ratio (e.g., 0.1), however, the contribution from lateral conduction is quite significant and needs to be included in a meaningful approximate solution. In this study, we derive closed form analytical expressions for the first order and the second order terms in the asymptotic expansion. The resulting asymptotic solution is capable of predicting the temperature distribution accurately including the effects of both incident angle and lateral heat conduction even at a moderate length scale ratio.

physics.class-ph↗
arXiv2026

Ion Implantation Enhanced Nucleation Facilitates Heat Transport across Atomically-Sharp Semiconductor Interfaces

Overheating is a critical bottleneck limiting the performance and reliability of next-generation high-power and high-frequency electronics. Interfacial thermal resistance constitutes a significant portion of the total thermal resistance. In this study, we report an ultrahigh thermal boundary conductance (TBC) of approximately 800 MW/m2-K at the atomically-sharp AlN-SiC interface, achieved through an ion implantation-enhanced nucleation epitaxy technique. This value is among the highest TBC values reported for semiconductor interfaces, confirmed by structural characterizations which show an ultrahigh-quality interface. Atomistic Green Function calculations reveal that elastic phonon transmission dominates the interface, with nearly half of the acoustic modes (0-15 THz) exhibiting near-unity transmission due to the atomically sharp structure. Furthermore, using high-energy-resolution electron energy loss spectroscopy, we probe vibrational properties with nanometer spatial resolution and identify unique interfacial phonon modes connecting the mismatched phonon spectra, confirmed by molecular dynamics simulations. The ultrahigh TBC is attributed to both the high elastic phonon transmission due to the high quality interfaces and the inelastic phonon scattering channel due to interfacial phonon modes. These findings not only advance the fundamental understanding of interfacial thermal transport but also provide a pathway for effective thermal management in emerging electronic devices.

cond-mat.mes-hall↗
arXiv2026

Research on Adaptive Inertial Control in Synchronization Systems: Based on Variational Optimization Methods and Their Applications in the Stability of Complex Networks

Aiming at the core problem that it is difficult for a fixed inertia coefficient to balance transient disturbance suppression and long-term stability in complex network synchronization systems, an adaptive inertia control strategy based on variational optimization is proposed. Taking the Kuramoto model with inertia as the research carrier, the analytical expression of the time-varying inertia coefficient M(t) is strictly derived by the functional variational method, and a hierarchical control structure of "benchmark inertia + disturbance feedback" is constructed to achieve the organic unity of minimizing the vulnerability performance function H(T) and stability constraints. A multimodal decoupling control strategy based on Laplacian eigenvector projection is designed to enhance the feedback strength of the dominant mode by eigenvalue weighting, improving the control accuracy and dynamic response speed. Simulation verification is carried out in complex network systems, and the control performance of regular networks (RG), random networks (ER), small-world networks (SW), scale-free networks (SF) and spider webs (SP) under three typical disturbances of pulses, monotonic decays and oscillatory decays is systematically analyzed. The results show that the proposed strategy reduces H(T) of the five networks by 19%-25%, shortens the relaxation time by 15%-24%, and the real parts of all system eigenvalues are less than -0.25s^-1 , meeting the asymptotic stability criterion. This study provides a new theoretical framework and engineering implementation scheme for the stability control of complex network synchronization systems, which can be widely applied to fields such as power grids, communication networks, and neural networks.

eess.SY↗
arXiv2026

Three-dimensional visualization of X-ray micro-CT with large-scale datasets: Efficiency and accuracy for real-time interaction

As Micro-CT technology continues to refine its characterization of material microstructures, industrial CT ultra-precision inspection is generating increasingly large datasets, necessitating solutions to the trade-off between accuracy and efficiency in the 3D characterization of defects during ultra-precise detection. This article provides a unique perspective on recent advances in accurate and efficient 3D visualization using Micro-CT, tracing its evolution from medical imaging to industrial non-destructive testing (NDT). Among the numerous CT reconstruction and volume rendering methods, this article selectively reviews and analyzes approaches that balance accuracy and efficiency, offering a comprehensive analysis to help researchers quickly grasp highly efficient and accurate 3D reconstruction methods for microscopic features. By comparing the principles of computed tomography with advancements in microstructural technology, this article examines the evolution of CT reconstruction algorithms from analytical methods to deep learning techniques, as well as improvements in volume rendering algorithms, acceleration, and data reduction. Additionally, it explores advanced lighting models for high-accuracy, photorealistic, and efficient volume rendering. Furthermore, this article envisions potential directions in CT reconstruction and volume rendering. It aims to guide future research in quickly selecting efficient and precise methods and developing new ideas and approaches for real-time online monitoring of internal material defects through virtual-physical interaction, for applying digital twin model to structural health monitoring (SHM).

cs.CV↗
arXiv2026

Spatial Multi-Task Learning for Breast Cancer Molecular Subtype Prediction from Single-Phase DCE-MRI

Accurate molecular subtype classification is essential for personalized breast cancer treatment, yet conventional immunohistochemical analysis relies on invasive biopsies and is prone to sampling bias. Although dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) enables non-invasive tumor characterization, clinical workflows typically acquire only single-phase post-contrast images to reduce scan time and contrast agent dose. In this study, we propose a spatial multi-task learning framework for breast cancer molecular subtype prediction from clinically practical single-phase DCE-MRI. The framework simultaneously predicts estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) status, and the Ki-67 proliferation index -- biomarkers that collectively define molecular subtypes. The architecture integrates a deep feature extraction network with multi-scale spatial attention to capture intratumoral and peritumoral characteristics, together with a region-of-interest weighting module that emphasizes the tumor core, rim, and surrounding tissue. Multi-task learning exploits biological correlations among biomarkers through shared representations with task-specific prediction branches. Experiments on a dataset of 960 cases (886 internal cases split 7:1:2 for training/validation/testing, and 74 external cases evaluated via five-fold cross-validation) demonstrate that the proposed method achieves an AUC of 0.893, 0.824, and 0.857 for ER, PR, and HER2 classification, respectively, and a mean absolute error of 8.2\% for Ki-67 regression, significantly outperforming radiomics and single-task deep learning baselines. These results indicate the feasibility of accurate, non-invasive molecular subtype prediction using standard imaging protocols.

cs.CV↗
arXiv2026

TokenSeg: Efficient 3D Medical Image Segmentation via Hierarchical Visual Token Compression

Three-dimensional medical image segmentation is a fundamental yet computationally demanding task due to the cubic growth of voxel processing and the redundant computation on homogeneous regions. To address these limitations, we propose \textbf{TokenSeg}, a boundary-aware sparse token representation framework for efficient 3D medical volume segmentation. Specifically, (1) we design a \emph{multi-scale hierarchical encoder} that extracts 400 candidate tokens across four resolution levels to capture both global anatomical context and fine boundary details; (2) we introduce a \emph{boundary-aware tokenizer} that combines VQ-VAE quantization with importance scoring to select 100 salient tokens, over 60\% of which lie near tumor boundaries; and (3) we develop a \emph{sparse-to-dense decoder} that reconstructs full-resolution masks through token reprojection, progressive upsampling, and skip connections. Extensive experiments on a 3D breast DCE-MRI dataset comprising 960 cases demonstrate that TokenSeg achieves state-of-the-art performance with 94.49\% Dice and 89.61\% IoU, while reducing GPU memory and inference latency by 64\% and 68\%, respectively. To verify the generalization capability, our evaluations on MSD cardiac and brain MRI benchmark datasets demonstrate that TokenSeg consistently delivers optimal performance across heterogeneous anatomical structures. These results highlight the effectiveness of anatomically informed sparse representation for accurate and efficient 3D medical image segmentation.

cs.CV↗
arXiv2025

Reinforcement Learning for Large Model: A Survey

Recent advances at the intersection of reinforcement learning (RL) and visual intelligence have enabled agents that not only perceive complex visual scenes but also reason, generate, and act within them. This survey offers a critical and up-to-date synthesis of the field. We first formalize visual RL problems and trace the evolution of policy-optimization strategies from RLHF to verifiable reward paradigms, and from Proximal Policy Optimization to Group Relative Policy Optimization. We then organize more than 200 representative works into four thematic pillars: multi-modal large language models, visual generation, unified model frameworks, and vision-language-action models. For each pillar we examine algorithmic design, reward engineering, benchmark progress, and we distill trends such as curriculum-driven training, preference-aligned diffusion, and unified reward modeling. Finally, we review evaluation protocols spanning set-level fidelity, sample-level preference, and state-level stability, and we identify open challenges that include sample efficiency, generalization, and safe deployment. Our goal is to provide researchers and practitioners with a coherent map of the rapidly expanding landscape of visual RL and to highlight promising directions for future inquiry. Resources are available at: https://github.com/weijiawu/Awesome-Visual-Reinforcement-Learning.

cs.CV↗
arXiv2025

EcomBench: Towards Holistic Evaluation of Foundation Agents in E-commerce

Foundation agents have rapidly advanced in their ability to reason and interact with real environments, making the evaluation of their core capabilities increasingly important. While many benchmarks have been developed to assess agent performance, most concentrate on academic settings or artificially designed scenarios while overlooking the challenges that arise in real applications. To address this issue, we focus on a highly practical real-world setting, the e-commerce domain, which involves a large volume of diverse user interactions, dynamic market conditions, and tasks directly tied to real decision-making processes. To this end, we introduce EcomBench, a holistic E-commerce Benchmark designed to evaluate agent performance in realistic e-commerce environments. EcomBench is built from genuine user demands embedded in leading global e-commerce ecosystems and is carefully curated and annotated through human experts to ensure clarity, accuracy, and domain relevance. It covers multiple task categories within e-commerce scenarios and defines three difficulty levels that evaluate agents on key capabilities such as deep information retrieval, multi-step reasoning, and cross-source knowledge integration. By grounding evaluation in real e-commerce contexts, EcomBench provides a rigorous and dynamic testbed for measuring the practical capabilities of agents in modern e-commerce.

cs.AI↗
arXiv2025

Optimal Interference Signal for Masking an Acoustic Source

In an environment where acoustic privacy or deliberate signal obfuscation is desired, it is necessary to mask the acoustic signature generated in essential operations. We consider the problem of masking the effect of an acoustic source in a target region where possible detection sensors are located. Masking is achieved by placing interference signals near the acoustic source. We introduce a theoretical and computational framework for designing such interference signals with the goal of minimizing the residual amplitude in the target region. For the three-dimensional (3D) forced wave equation with spherical symmetry, we derive analytical quasi-steady periodic solutions for several canonical cases. We examine the phenomenon of self-masking where an acoustic source with certain spatial forcing profile masks itself from detection outside its forcing footprint. We then use superposition of spherically symmetric solutions to investigate masking in a given target region. We analyze and optimize the performance of using one or two point-forces deployed near the acoustic source for masking in the target region. For the general case where the spatial forcing profile of the acoustic source lacks spherical symmetry, we develop an efficient numerical method for solving the 3D wave equation. Potential applications of this work include undersea acoustic communication security, undersea vehicles stealth, and protection against acoustic surveillance.

math.AP↗
arXiv2025

ZipAR: Parallel Auto-regressive Image Generation through Spatial Locality

In this paper, we propose ZipAR, a training-free, plug-and-play parallel decoding framework for accelerating auto-regressive (AR) visual generation. The motivation stems from the observation that images exhibit local structures, and spatially distant regions tend to have minimal interdependence. Given a partially decoded set of visual tokens, in addition to the original next-token prediction scheme in the row dimension, the tokens corresponding to spatially adjacent regions in the column dimension can be decoded in parallel, enabling the ``next-set prediction'' paradigm. By decoding multiple tokens simultaneously in a single forward pass, the number of forward passes required to generate an image is significantly reduced, resulting in a substantial improvement in generation efficiency. Experiments demonstrate that ZipAR can reduce the number of model forward passes by up to 91% on the Emu3-Gen model without requiring any additional retraining. Code is available here: https://github.com/ThisisBillhe/ZipAR.

cs.CV↗