arXiv subjects

Jia Zhou

Publications and source records attributed to Jia Zhou.

At least 37 records · Page 2Linked to original sources

arXiv2025

Semi-Degree Condition for Arbitrary $H$-Linked Oriented Graphs

Let $ H $ be a multi-digraph on $ h $ vertices with $ q $ arcs. An \textbf{$H$-subdivision} in a digraph $D$ is a subdigraph obtained by replacing every arc $uv$ of $H$ with a path from $u$ to $v$ in $D$ such that these paths are pairwise internally vertex-disjoint. A digraph $ D $ is \textbf{arbitrary $ H $-linked} if, for every injection $ f: V(H) \to V(D) $, there exists an $ H $-subdivision in $ D $ such that each vertex $ v \in V(H) $ is mapped to $ f(v) \in V(D) $, and the length of every subdivision path can be arbitrarily specified as {an integer $l \geq 4$}. An oriented graph is a digraph without 2-cycles. Keevash, Kühn, and Osthus proved that every sufficiently large oriented graph $ D $ of order $ n $ with $δ^0(D) \geq \frac{3n-4}{8}$ contains a Hamilton cycle (i.e., a $\overset{\leftrightarrow}{K_2}$-subdivision). Subsequently, Kelly, Kühn, and Osthus showed that such oriented graphs {are also arbitrary $ H $-linked, where $H$ is a loop}. Motivated by these results, we establish a minimum semi-degree condition for arbitrary $ H $-linked oriented graphs: there exists $ n_0 = n_0(h,q) $ such that every oriented graph $ D $ of order $ n \geq n_0 $ with $δ^0(D) \geq \frac{3n + 3h + 3q - 5}{8}$ is arbitrary $ H $-linked; specifically, if $H$ is a loop, this holds under the weaker condition $δ^0(D) \geq \frac{3n - 4}{8}$. The result provides an oriented graph analogue of Wang's conjecture on cycle-factors in graphs [J. Korean Math. Soc. 51 (2014) 919--940] and determines the tight semi-degree bounds for both strongly Hamiltonian-connected and arbitrary $q$-linked oriented graphs.

math.CO↗

arXiv2025

Testing Risk Difference of Two Proportions for Combined Unilateral and Bilateral Data

In clinical studies with paired organs, binary outcomes often exhibit intra-subject correlation and may include a mixture of unilateral and bilateral observations. Under Donner's constant correlation model, we develop three likelihood-based test statistics (the likelihood ratio, Wald-type, and score tests) for assessing the risk difference between two proportions. Simulation studies demonstrate good control of type I error and comparable power among the three tests, with the score test showing slightly better stability. Applications to otolaryngologic and ophthalmologic data illustrate the methods. An online calculator is also provided for power analysis and risk difference testing. The score test is recommended for practical use and future studies with combined unilateral and bilateral binary data.

stat.ME↗

arXiv2025

ERSR: An Ellipse-constrained pseudo-label refinement and symmetric regularization framework for semi-supervised fetal head segmentation in ultrasound images

Automated segmentation of the fetal head in ultrasound images is critical for prenatal monitoring. However, achieving robust segmentation remains challenging due to the poor quality of ultrasound images and the lack of annotated data. Semi-supervised methods alleviate the lack of annotated data but struggle with the unique characteristics of fetal head ultrasound images, making it challenging to generate reliable pseudo-labels and enforce effective consistency regularization constraints. To address this issue, we propose a novel semi-supervised framework, ERSR, for fetal head ultrasound segmentation. Our framework consists of the dual-scoring adaptive filtering strategy, the ellipse-constrained pseudo-label refinement, and the symmetry-based multiple consistency regularization. The dual-scoring adaptive filtering strategy uses boundary consistency and contour regularity criteria to evaluate and filter teacher outputs. The ellipse-constrained pseudo-label refinement refines these filtered outputs by fitting least-squares ellipses, which strengthens pixels near the center of the fitted ellipse and suppresses noise simultaneously. The symmetry-based multiple consistency regularization enforces multi-level consistency across perturbed images, symmetric regions, and between original predictions and pseudo-labels, enabling the model to capture robust and stable shape representations. Our method achieves state-of-the-art performance on two benchmarks. On the HC18 dataset, it reaches Dice scores of 92.05% and 95.36% with 10% and 20% labeled data, respectively. On the PSFH dataset, the scores are 91.68% and 93.70% under the same settings.

cs.CV↗

arXiv2025

Homogeneity Test of Proportions for Combined Unilateral and Bilateral Data via GEE and MLE Approaches

In clinical trials involving paired organs such as eyes, ears, and kidneys, binary outcomes may be collected bilaterally or unilaterally. In such combined datasets, bilateral outcomes exhibit intra-subject correlation, while unilateral outcomes are assumed independent. We investigate the generalized Estimating Equations (GEE) approach for testing homogeneity of proportions across multiple groups for the combined unilateral and bilateral data, and compare it with three likelihood-based statistics (likelihood ratio, Wald-type, and score) under Rosner's constant $R$ model and Donner's equal correlation $ρ$ model. Monte Carlo simulations evaluate empirical type I error and power under varied sample sizes and parameter settings. The GEE and score tests show superior type I error control, outperforming likelihood ratio and Wald-type tests. Applications to two real datasets in otolaryngologic and ophthalmologic studies illustrate the methods. We recommend the GEE and score tests for homogeneity testing, and suggest GEE for more complex models with covariates, while favoring the score statistic for small sample exact tests due to its computational efficiency.

stat.ME↗

arXiv2025

Mitigating Modality Quantity and Quality Imbalance in Multimodal Online Federated Learning

The Internet of Things (IoT) ecosystem produces massive volumes of multimodal data from diverse sources, including sensors, cameras, and microphones. With advances in edge intelligence, IoT devices have evolved from simple data acquisition units into computationally capable nodes, enabling localized processing of heterogeneous multimodal data. This evolution necessitates distributed learning paradigms that can efficiently handle such data. Furthermore, the continuous nature of data generation and the limited storage capacity of edge devices demand an online learning framework. Multimodal Online Federated Learning (MMO-FL) has emerged as a promising approach to meet these requirements. However, MMO-FL faces new challenges due to the inherent instability of IoT devices, which often results in modality quantity and quality imbalance (QQI) during data collection. In this work, we systematically investigate the impact of QQI within the MMO-FL framework and present a comprehensive theoretical analysis quantifying how both types of imbalance degrade learning performance. To address these challenges, we propose the Modality Quantity and Quality Rebalanced (QQR) algorithm, a prototype learning based method designed to operate in parallel with the training process. Extensive experiments on two real-world multimodal datasets show that the proposed QQR algorithm consistently outperforms benchmarks under modality imbalance conditions with promising learning performance.

cs.LG↗

arXiv2025

Proof of the linkage conjecture for highly connected tournaments

A digraph $D$ is $k$-linked if for every $2k$ distinct vertices $ x_1,\ldots , x_k, y_1, \ldots , y_k$ in $D$, there exist $k$ pairwise vertex-disjoint paths $P_1,\ldots, P_k$ such that $P_i$ starts at $x_i$ and ends at $y_i$ for each $i\in [k]$. In 2021, Girão, Popielarz, and Snyder [Combinatorica 41 (2021) 815--837] conjectured that there exists a constant $C >0$ such that every $(2k+1)$-connected tournament with minimum out-degree at least $Ck$ is $k$-linked. In this paper, we disprove this conjecture by constructing a family of counterexamples with minimum out-degree at least $\frac{k^2+11k}{26}$ (for $k\geq 42$). Further, we prove that every $(2k+1)$-connected semicomplete digraph $D$ with minimum out-degree at least $ 7k^2 + 36k$ is $k$-linked. This result is optimal in terms of both connectivity and minimum out-degree (up to a multiplicative factor), which refines and generalizes the earlier result of Girão, Popielarz, and Snyder.

math.CO↗

arXiv2025

Goodness-of-fit Tests for Combined Unilateral and Bilateral Data

Clinical trials involving paired organs often yield a mixture of unilateral and bilateral data, where each subject may contribute either one or two responses under certain circumstances. While unilateral responses from different individuals can be treated as independent, bilateral responses from the same individual are likely correlated. Various statistical methods have been developed to account for this intra-subject correlation in the bilateral data, and in practice it is crucial to select an appropriate model for accurate inference. Tang et. al. (2012) discussed model selection issues using a variety of goodness-of-fit test statistics for correlated bilateral data for two groups, and Liu and Ma (2020) extended these methods to settings with $g\ge2$ groups. In this work, we investigate the goodness-of-fit statistics for the combined unilateral and bilateral data under different statistical models that address the intra-subject correlation, including the Clayton copula model, in addition to those considered in prior studies. Simulation results indicate that the performance of the goodness-of-fit tests is model-dependent, especially when the sample size is small and/or the intra-subject correlation is high, which is consistent with the findings by Liu and Ma (2020) for purely bilateral data. Applications to real data from otolaryngologic and ophthalmologic studies are included.

stat.ME↗

arXiv2025

Generalized derivations of Complex $ω$-Lie Superalgebras

~Let $(g,~[-,-],~ω)$ be a finite-dimensional complex $ω$-Lie superalgebra. This paper explores the algbaraic structures of generalized derivation superalgebra ${\rm GDer}(g)$, compatatible generalized derivations algebra ${\rm GDer}^ω(g)$, and their subvarieties such as quasiderivation superalgebra ${\rm QDer}(g)$(${\rm QDer}^ω(g)$), centroid ${\rm Cent}(g)$ (${\rm Cent}^ω(g)$) and quasicentroid ${\rm QCent}(g)$ (${\rm QCent}^ω(g)$). We prove that ${\rm GDer}^ω(g) = {\rm QDer}^ω(g) + {\rm QCent}^ω(g)$. We also study the embedding question of compatible quasiderivations of $ω$-Lie superalgebras, demonstrating that ${\rm QDer}^ω(g)$ can be embedded as derivations in a larger $ω$-Lie superalgebra $\breve g$ and furthermore, we obtain a semidirect sum decomposition: ${\rm Der}^ω(\breve{g})=φ({\rm QDer}^ω(g))\oplus {\rm ZDer}(\breve{g})$, when the annihilator of $g$ is zero. In particular, for the 3-dimensional complex $ω$-Lie superalgebra $H$, we explicitly calculate ${\rm GDer}(H)$, ${\rm GDer}^ω(H)$, ${\rm QDer}(H)$ and ${\rm QDer}^ω(H)$, and derive the Jordan standard forms of generic elements in these varieties.

math.RA↗

arXiv2025

Computation and Communication Efficient Lightweighting Vertical Federated Learning for Smart Building IoT

With the increasing number and enhanced capabilities of IoT devices in smart buildings, these devices are evolving beyond basic data collection and control to actively participate in deep learning tasks. Federated Learning (FL), as a decentralized learning paradigm, is well-suited for such scenarios. However, the limited computational and communication resources of IoT devices present significant challenges. While existing research has extensively explored efficiency improvements in Horizontal FL, these techniques cannot be directly applied to Vertical FL due to fundamental differences in data partitioning and model structure. To address this gap, we propose a Lightweight Vertical Federated Learning (LVFL) framework that jointly optimizes computational and communication efficiency. Our approach introduces two distinct lightweighting strategies: one for reducing the complexity of the feature model to improve local computation, and another for compressing feature embeddings to reduce communication overhead. Furthermore, we derive a convergence bound for the proposed LVFL algorithm that explicitly incorporates both computation and communication lightweighting ratios. Experimental results on an image classification task demonstrate that LVFL effectively mitigates resource demands while maintaining competitive learning performance.

cs.LG↗

arXiv2025

Quantization of Visible Light by a Ni$_2$ Molecular Optical Resonator

The quantization of an optical field is a frontier in quantum optics with implications for both fundamental science and technological applications. Here, we demonstrate that a dinickel complex (Ni$_2$) traps and quantizes classical visible light, behaving as an individual quantum system or the Jaynes Cummings molecule.The composite system forms through coherently coupling the two level NiNi charge transfer transition with the local scattering field, which produces nonclassical light featuring photon anti bunching and squeezed states, as verified by a sequence of discrete photonic modes in the incoherent resonance fluorescence. Notably, in this Ni$_2$ system, the collective coupling of N molecule ensembles scales as N, distinct from the Tavis-Cummings model, which allows easy achievement of ultrastrong coupling. This is exemplified by a vacuum Rabi splitting of 1.2 eV at the resonance (3.25 eV) and a normalized coupling rate of 0.18 for the N = 4 ensemble. The resulting quantum light of single photonic modes enables driving the molecule field interaction in cavity free solution, which profoundly modifies the electronic states. Our results establish Ni$_2$ as a robust platform for quantum optical phenomena under ambient conditions, offering new pathways for molecular physics, polaritonic chemistry and quantum information processing.

quant-ph↗

arXiv2025

Quadruply Bonded Mo$_2$ Molecules Acting as an Inborn Emitter-Resonator Quantum System in Free Space

In recent decades, significant progress has been made in construction and study of individual quantum systems consisting of the basic single matter and energy particles, i.e., atoms and photons, which show great potentials in quantum computation and communication. Here, we demonstrate that the quadruply-bonded Mo$_2$ unit of the complex can trap photons of visible light under ambient conditions, producing intense local electromagnetic (EM) field that features squeezed states, photon antibunching, and vacuum Rabi splitting. Our results show that both the electronic and vibrational states of the Mo$_2$ molecule are modified by coherent coupling with the scattered photons of the Mo$_2$ unit, as evidenced by the Rabi doublet4 and the Mollow triplet in the incoherent resonance fluorescence and the Raman spectra. The Mo$_2$ molecule, acting as an independent emitter-resonator integrated quantum system, allows optical experiments to be conducted in free space, enabling fundamental quantum phenomena to be observed through conventional spectroscopic instrumentation. This provides a new platform for study of field effects and quantum electrodynamics (QED) in the optical domain. The insights gained from this study advance our understanding in metal-metal bond chemistry, molecular physics and quantum optics, with applications in quantum information processing, optoelectronic devices and control of chemical reactivity.

quant-ph↗

arXiv2025

AIstorian lets AI be a historian: A KG-powered multi-agent system for accurate biography generation

Huawei has always been committed to exploring the AI application in historical research. Biography generation, as a specialized form of abstractive summarization, plays a crucial role in historical research but faces unique challenges that existing large language models (LLMs) struggle to address. These challenges include maintaining stylistic adherence to historical writing conventions, ensuring factual fidelity, and handling fragmented information across multiple documents. We present AIstorian, a novel end-to-end agentic system featured with a knowledge graph (KG)-powered retrieval-augmented generation (RAG) and anti-hallucination multi-agents. Specifically, AIstorian introduces an in-context learning based chunking strategy and a KG-based index for accurate and efficient reference retrieval. Meanwhile, AIstorian orchestrates multi-agents to conduct on-the-fly hallucination detection and error-type-aware correction. Additionally, to teach LLMs a certain language style, we finetune LLMs based on a two-step training approach combining data augmentation-enhanced supervised fine-tuning with stylistic preference optimization. Extensive experiments on a real-life historical Jinshi dataset demonstrate that AIstorian achieves a 3.8x improvement in factual accuracy and a 47.6% reduction in hallucination rate compared to existing baselines. The data and code are available at: https://github.com/ZJU-DAILY/AIstorian.

cs.CL↗

arXiv2025

On the $k$-linkage problem for generalizations of semicomplete digraphs

A directed graph (digraph) $ D $ is $ k $-linked if $ |D| \geq 2k $, and for any $ 2k $ distinct vertices $ x_1, \ldots, x_k, y_1, \ldots, y_k $ of $ D $, there exist vertex-disjoint paths $ P_1, \ldots, P_k $ such that $ P_i $ is a path from $ x_i $ to $ y_i $ for each $ i \in [k] $. In 1980, Thomassen conjectured that there exists a function $ f(k) $ such that every $ f(k) $-strong digraph is $ k $-linked. He later disproved this conjecture by showing that $ f(2) $ does not exist for general digraphs and proved that the function $f(k)$ exists for the class of tournaments. In this paper we consider a large class $\mathcal{ D} $ of digraphs which includes all semicomplete digraphs (digraphs with no pair of non-adjacent vertices) and all quasi-transitive digraphs (a digraph $D$ is quasi-transitive if for any three vertices $x, y, z$ of $D$, whenever $xy$ and $yz$ are arcs, then $x$ and $z$ are adjacent). We prove that every $ 3k $-strong digraph $D\in \mathcal{D}$ with minimum out-degree at least $ 23k $ is $ k $-linked. A digraph $D$ is $l$-quasi-transitive if whenever there is a path of length $l$ between vertices $u$ and $v$ in $D$ the vertices $u$ and $v$ are adjacent. Hence 2-quasi-transitive digraphs are exactly the quasi-transitive digraphs. We prove that there is a function $f(k,l)$ so that every $f(k,l)$-strong $l$-quasi-transitive digraph is $k$-linked. The main new tool in our proofs significantly strengthens an important property of vertices with maximum in-degree in a tournament. While Landau in 1953 already proved that such a vertex $v$ is reachable by all other vertices by paths of length at most 2, we show that, in fact, the structure of these paths is much richer. In general there are many such paths for almost all out-neighbours of $v$ and this property is crucial in our proofs.

math.CO↗

arXiv2025

Beam test result and digitization of TaichuPix-3: A Monolithic Active Pixel Sensors for CEPC vertex detector

The Circular Electron-Positron Collider (CEPC), as the next-generation electron-positron collider, is tasked with advancing not only Higgs physics but also the discovery of new physics. Achieving these goals requires high-precision measurements of particles. Taichu seires, Monolithic Active Pixel Sensor (MAPS), a key component of the vertex detector for CEPC was designed to meet the CEPC's requirements. For the geometry of vertex detector is long barrel with no endcap, and current silicon lacks a complete digitization model, precise estimation of cluster size particularly causing by particle with large incident angle is needed. Testbeam results were conducted at the Beijing Synchrotron Radiation Facility (BSRF) to evaluate cluster size dependence on different incident angles and threshold settings. Experimental results confirmed that cluster size increases with incident angle. Simulations using the Allpix$^2$ framework replicated experimental trends at small angles but exhibited discrepancies at large angles, suggesting limitations in linear electric field assumptions and sensor thickness approximations. The results from both testbeam and simulations have provided insights into the performance of the TaichuPix chip at large incident angles, offering a crucial foundation for the establishment of a digital model and addressing the estimation of cluster size in the forward region of the long barrel. Furthermore, it offers valuable references for future iterations of TaichuPix, the development of digital models, and the simulation and estimation of the vertex detector's performance.

physics.ins-det↗

arXiv2024

Solutions to the linkage conjecture in tournaments

A digraph $D$ is $k$-linked if for every $2k$-tuple $ x_1,\ldots , x_k, y_1, \ldots , y_k$ of distinct vertices in $D$, there exist $k$ pairwise vertex-disjoint paths $P_1,\ldots, P_k$ such that $P_i$ starts at $x_i$ and ends at $y_i$, $i\in [k]$. In 2015, Pokrovskiy conjectured that there exists a function $g(k)$ such that every $2k$-connected tournament with minimum in-degree and minimum out-degree at least $g(k)$ is $k$-linked in [J. Comb. Theory, Ser. B 115 (2015) 339--347]. In this paper, we disprove this conjecture by constructing a family of counterexamples. The counterexamples also provide a negative answer to the question raised by Girão, Popielarz, Snyder in [Combinatorica 41 (2021) 815--837]. Further, we prove that every $(2k+1)$-connected semicomplete digraph $D$ with minimum out-degree at least $ 10^7k^{4}$ is $k$-linked, which refines and generalizes the early result of Girão, Popielarz, Snyder.

math.CO↗

arXiv2024

New Physics Off the $Z$-Pole: $e^+ e^- \rightarrow f \bar f$ at Future Lepton Colliders

We explore the prospects for probing new physics (NP) beyond the Standard Model (SM) at future lepton colliders through precision measurements of $e^+e^-\to f{\bar f}$ observables off the $Z$ resonance. We consider interference between SM contributions and those arising from dimension-6, four-fermion effective operators that encode the effects of NP, yielding a linear dependence on the latter. This linear dependence in general increases with magnitude of the collision energy offset from the $Z$ pole. We consider a variety of asymmetries in order to enhance the NP-sensitivity while reducing experimental systematic and theoretical, SM uncertainties: an inclusive above and below $Z$-resonance total cross section asymmetry ($A_σ$) as well as the conventional forward-backward ($A_{\rm FB}$) and polarization ($A_{\rm pol}$) asymmetries. Based on projected statistical uncertainties at the Circular Electron-Positron Collider (CEPC), we find that t measurement of $A_σ$ could extend the sensitivity to the NP mass scale by as much as a factor of $\sim 7$ compared to the present reach obtained with the CERN Large Electron Positron Collider. Inclusion of projected systematic theoretical SM uncertainties substantially reduce this sensitivity gain. For $A_{\rm FB}$, inclusion of experimental systematic uncertainties has a marginal impact on the gain in NP reach, whereas SM theoretical uncertainties remain a significant barrier to realizing the full NP sensitivity. Analogous conclusions apply to the CERN Future Circular Collider (FCC-ee) and International Linear Collider (ILC).

hep-ph↗

arXiv2024

Implementation of two-dimensional selective acoustic tweezers merely using four straight interdigitated transducers:a numerical proof of concept of radiation field synthesis by pulsed acoustic waves

Selective acoustic tweezers can focus the acoustic radiation force on a single particle to manipulate it without affecting its neighbors. This has long required highly complex hardware. In this numerical study, we show that pulsed acoustic waves can be used for the selective manipulation of particles using only two pairs of orthogonal transducers. While these tweezers are well-known for their ability to manipulate arrays of particles, we show that selectivity can be achieved by using sequences of acoustic pulses to iteratively construct a combined acoustic potential focused only on the target particle.

physics.flu-dyn↗

arXiv2024

Performance study of the JadePix-3 telescope from a beam test

We present the results of a beam test conducted on a telescope using the JadePix-3 pixel sensor, developed with TowerJazz 180 nm CMOS imaging technology. The telescope is composed of five planes, each equipped with a JadePix-3 sensor with pitches of 26 um x 16 um and 23.11 um x 16 um. In addition, it features an FPGA-based synchronous readout system. The telescope underwent testing using an electron beam with energy ranging from 4 to 6 GeV. At an electron energy of 5.4 GeV, the telescope demonstrated superior spatial resolutions of 2.6 and 2.3 um in two dimensions. By designating the central plane as the device under test, we evaluated the JadePix-3 sensor's spatial resolutions as 5.2 and 4.6 um in two dimensions, achieving a detection efficiency of over 99%.

hep-ex↗