arXiv subjects

Xiaoyu Zhang

Publications and source records attributed to Xiaoyu Zhang.

At least 91 records · Page 5Linked to original sources

arXiv2025

YingMusic-SVC: Real-World Robust Zero-Shot Singing Voice Conversion with Flow-GRPO and Singing-Specific Inductive Biases

Singing voice conversion (SVC) aims to render the target singer's timbre while preserving melody and lyrics. However, existing zero-shot SVC systems remain fragile in real songs due to harmony interference, F0 errors, and the lack of inductive biases for singing. We propose YingMusic-SVC, a robust zero-shot framework that unifies continuous pre-training, robust supervised fine-tuning, and Flow-GRPO reinforcement learning. Our model introduces a singing-trained RVC timbre shifter for timbre-content disentanglement, an F0-aware timbre adaptor for dynamic vocal expression, and an energy-balanced rectified flow matching loss to enhance high-frequency fidelity. Experiments on a graded multi-track benchmark show that YingMusic-SVC achieves consistent improvements over strong open-source baselines in timbre similarity, intelligibility, and perceptual naturalness, especially under accompanied and harmony-contaminated conditions, demonstrating its effectiveness for real-world SVC deployment.

cs.SD↗

arXiv2025

Adaptive Legged Locomotion via Online Learning for Model Predictive Control

We provide an algorithm for adaptive legged locomotion via online learning and model predictive control. The algorithm is composed of two interacting modules: model predictive control (MPC) and online learning of residual dynamics. The residual dynamics can represent modeling errors and external disturbances. We are motivated by the future of autonomy where quadrupeds will autonomously perform complex tasks despite real-world unknown uncertainty, such as unknown payload and uneven terrains. The algorithm uses random Fourier features to approximate the residual dynamics in reproducing kernel Hilbert spaces. Then, it employs MPC based on the current learned model of the residual dynamics. The model is updated online in a self-supervised manner using least squares based on the data collected while controlling the quadruped. The algorithm enjoys sublinear \textit{dynamic regret}, defined as the suboptimality against an optimal clairvoyant controller that knows how the residual dynamics. We validate our algorithm in Gazebo and MuJoCo simulations, where the quadruped aims to track reference trajectories. The Gazebo simulations include constant unknown external forces up to $12\boldsymbol{g}$, where $\boldsymbol{g}$ is the gravity vector, in flat terrain, slope terrain with $20\degree$ inclination, and rough terrain with $0.25m$ height variation. The MuJoCo simulations include time-varying unknown disturbances with payload up to $8~kg$ and time-varying ground friction coefficients in flat terrain.

cs.RO↗

arXiv2025

Noncollinear Spin-Flip TDDFT for Potential Energy Surface Crossings: Conical Intersections and Spin Crossings

We recently proposed a scheme to generalize collinear functionals to the noncollinear regime, termed the multicollinear approach. The resulting noncollinear functionals preserve spin symmetry while providing numerically stable higher-order functional derivatives. This scheme has already been applied to noncollinear spin-flip TDDFT and its analytic gradient calculations. In the present work, with the aid of the penalty function method, we employ the noncollinear spin-flip TDDFT in multicollinear scheme to locate potential energy surface crossings. We investigate two distinct types of crossings and analyze their topographical and spin characteristics near the crossing points. The first type is conical intersections, typically involving two singlet states such as the ground and first excited states. The second type involves spin crossings that occur between electronic states with different spin multiplicities, such as between singlet and triplet. These crossing regions enable ultrafast nonadiabatic transitions through either nonadiabatic coupling or spin-orbit coupling, playing a crucial role in photochemistry. Through theoretical analysis and illustrative examples, we demonstrate the advantages of noncollinear spin-flip TDDFT over conventional collinear spin-flip TDDFT or spin-conserving TDDFT. Finally, we systematically evaluate its prospects as an electronic structure method for use in nonadiabatic molecular dynamics.

physics.chem-ph↗

arXiv2025

A General Highly Accurate Online Planning Method Integrating Large Language Models into Nested Rollout Policy Adaptation for Dialogue Tasks

In goal-oriented dialogue tasks, the main challenge is to steer the interaction towards a given goal within a limited number of turns. Existing approaches either rely on elaborate prompt engineering, whose effectiveness is heavily dependent on human experience, or integrate policy networks and pre-trained policy models, which are usually difficult to adapt to new dialogue scenarios and costly to train. Therefore, in this paper, we present Nested Rollout Policy Adaptation for Goal-oriented Dialogue (NRPA-GD), a novel dialogue policy planning method that completely avoids specific model training by utilizing a Large Language Model (LLM) to simulate behaviors of user and system at the same time. Specifically, NRPA-GD constructs a complete evaluation mechanism for dialogue trajectories and employs an optimization framework of nested Monte Carlo simulation and policy self-adaptation to dynamically adjust policies during the dialogue process. The experimental results on four typical goal-oriented dialogue datasets show that NRPA-GD outperforms both existing prompt engineering and specifically pre-trained model-based methods. Impressively, NRPA-GD surpasses ChatGPT and pre-trained policy models with only a 0.6-billion-parameter LLM. The proposed approach further demonstrates the advantages and novelty of employing planning methods on LLMs to solve practical planning tasks.

cs.CL↗

arXiv2025

Biomolecular LQR under Partial Observation

This paper introduces a biomolecular Linear Quadratic Regulator (LQR) to investigate the design principles of gene regulatory networks. We show that for fundamental gene regulation network, the bio-controller derived from LQR theory precisely recapitulate natural network motifs, such as auto-regulation and incoherent feedforward loops. This emulation arises from a fundamental principle: the LQR cost function mathematically encodes environmental survival demands, which subsequently drives the selection of both network topology and biochemical parameters. Our work thus establishes a theoretical basis for interpreting biological circuit design, directly linking evolutionary pressures to observable regulatory structures.

q-bio.MN↗

arXiv2025

Human Mobility in Epidemic Modeling

Human mobility forms the backbone of contact patterns through which infectious diseases propagate, fundamentally shaping the spatio-temporal dynamics of epidemics and pandemics. While traditional models are often based on the assumption that all individuals have the same probability of infecting every other individual in the population, a so-called random homogeneous mixing, they struggle to capture the complex and heterogeneous nature of real-world human interactions. Recent advancements in data-driven methodologies and computational capabilities have unlocked the potential of integrating high-resolution human mobility data into epidemic modeling, significantly improving the accuracy, timeliness, and applicability of epidemic risk assessment, contact tracing, and intervention strategies. This review provides a comprehensive synthesis of the current landscape in human mobility-informed epidemic modeling. We explore diverse sources and representations of human mobility data, and then examine the behavioral and structural roles of mobility and contact in shaping disease transmission dynamics. Furthermore, the review spans a wide range of epidemic modeling approaches, ranging from classical compartmental models to network-based, agent-based, and machine learning models. And we also discuss how mobility integration enhances risk management and response strategies during epidemics. By synthesizing these insights, the review can serve as a foundational resource for researchers and practitioners, bridging the gap between epidemiological theory and the dynamic complexities of human interaction while charting clear directions for future research.

cs.SI↗

arXiv2025

Direct Numerical Simulations of Oxygen-Flame-Driven Deflagration-to-Detonation Transition in Type Ia Supernovae

We present direct numerical simulations demonstrating deflagration-to-detonation transition (DDT) driven by oxygen flames in Type Ia supernova progenitors. Using the Castro hydrodynamics code coupled with the ``aprox13'' 13-isotope nuclear network, we simulate combustion in isolated fuel regions where oxygen flames trail carbon flames. In a fiducial one-dimensional run at $ρ_{0}=3.5\times10^{7}\ \mathrm{g\ cm^{-3}}$ we observe spontaneous DDT of the oxygen flame via the Zel'dovich gradient mechanism when the carbon-oxygen separation reaches $\sim 10\ \mathrm{km}$. The oxygen detonation then captures the carbon flame and triggers a stable carbon detonation. Systematic one-dimensional parameter scans show that successful carbon DDT requires upstream densities in the range $(3.1$--$3.6)\times10^{7}\ \mathrm{g\ cm^{-3}}$ and a minimum carbon-flame thickness of $\gtrsim 20\ \mathrm{m}$. Two-dimensional simulations confirm DDT and demonstrate that the multidimensional cellular structure of the oxygen detonation can promote carbon detonation at somewhat lower densities than in one dimension. These results provide direct numerical evidence that oxygen-flame-driven DDT is physically plausible in turbulent white-dwarf environments and underscore the importance of multidimensional effects for Type Ia supernova explosion modeling.

astro-ph.HE↗

arXiv2025

CITADEL: Context Similarity Based Deep Learning Framework Bug Finding

With the application of deep learning technology, tools of DL framework testing are in high demand. Existing DL framework testing tools have limited coverage of bug types. For example, they lack the capability of effectively finding performance bugs, which are critical for DL models regarding performance, economics, and the environment. Moreover, existing tools are inefficient, generating hundreds of test cases with few trigger bugs. In this paper, we propose Citadel, a method that accelerates bug finding in terms of efficiency and effectiveness. We observe that many DL framework bugs are similar due to the similarity of operators and algorithms belonging to the same family. Orthogonal to existing bug-finding tools, Citadel aims to find new bugs that are similar to reported ones that have known test oracles. Citadel defines context similarity to measure the similarity of DL framework API pairs and automatically generates test cases with oracles for APIs that are similar to the problematic APIs in existing bug reports. Citadel effectively detects 58 and 66 API bugs on PyTorch and TensorFlow (excluding those rejected by developers or duplicates of prior reports), many of which, e.g., 13 performance bugs, cannot be detected by existing tools. Moreover, 35.40% of test cases generated by Citadel can trigger bugs significantly transcending the state-of-the-art method (3.90%).

cs.SE↗

arXiv2025

DeepfakeBench-MM: A Comprehensive Benchmark for Multimodal Deepfake Detection

The misuse of advanced generative AI models has resulted in the widespread proliferation of falsified data, particularly forged human-centric audiovisual content, which poses substantial societal risks (e.g., financial fraud and social instability). In response to this growing threat, several works have preliminarily explored countermeasures. However, the lack of sufficient and diverse training data, along with the absence of a standardized benchmark, hinder deeper exploration. To address this challenge, we first build Mega-MMDF, a large-scale, diverse, and high-quality dataset for multimodal deepfake detection. Specifically, we employ 21 forgery pipelines through the combination of 10 audio forgery methods, 12 visual forgery methods, and 6 audio-driven face reenactment methods. Mega-MMDF currently contains 0.1 million real samples and 1.1 million forged samples, making it one of the largest and most diverse multimodal deepfake datasets, with plans for continuous expansion. Building on it, we present DeepfakeBench-MM, the first unified benchmark for multimodal deepfake detection. It establishes standardized protocols across the entire detection pipeline and serves as a versatile platform for evaluating existing methods as well as exploring novel approaches. DeepfakeBench-MM currently supports 5 datasets and 11 multimodal deepfake detectors. Furthermore, our comprehensive evaluations and in-depth analyses uncover several key findings from multiple perspectives (e.g., augmentation, stacked forgery). We believe that DeepfakeBench-MM, together with our large-scale Mega-MMDF, will serve as foundational infrastructures for advancing multimodal deepfake detection.

cs.CR↗

arXiv2025

High-dimensional low-rank matrix regression with unknown latent structures

We study low-rank matrix regression in settings where matrix-valued predictors and scalar responses are observed across multiple individuals. Rather than assuming a fully homogeneous coefficient matrices across individuals, we accommodate shared low-dimensional structure alongside individual-specific deviations. To this end, we introduce a tensor-structured homogeneity pursuit framework, wherein each coefficient matrix is represented as a product of shared low-rank subspaces and individualized low-rank loadings. We propose a scalable estimation procedure based on scaled gradient descent, and establish non-asymptotic bounds demonstrating that the proposed estimator attains improved convergence rates by leveraging shared information while preserving individual-specific signals. The framework is further extended to incorporate scaled hard thresholding for recovering sparse latent structures, with theoretical guarantees in both linear and generalized linear model settings. Our approach provides a principled middle ground between fully pooled and fully separate analyses, achieving strong theoretical performance, computational tractability, and interpretability in high-dimensional multi-individual matrix regression problems.

stat.ME↗

arXiv2025

Wave morphing and flat-top ground states in photonics systems driven by artificial gauge fields

In quantum physics, classical optics, and many other wave systems, wave confinement in a potential well is associated with discrete oscillatory states, and the ground state is typically assumed to vanish uniformly. An open question is whether the ground state can counterintuitively support a flat-top, nonzero envelope, offering new opportunities for quantum emitters, optical antennas, and lasers. Here, we show that by applying the Byers-Yang theorem with an artificial gauge field, energy levels can be continuously shifted, driving eigenstates to morph into a ground state with a uniform yet nontrivial wave envelope. We implement this concept in a photonic crystal slab where a central bulk region is surrounded by heterogeneous bandgaps that engineer reflective phases acting as an artificial local gauge field. By inducing lasing, we probe directly the evolution of the energy levels, demonstrating wave morphing toward a flat-top ground state via near- and far-field measurements.

physics.optics↗

arXiv2025

Synthetic Series-Symbol Data Generation for Time Series Foundation Models

Foundation models for time series analysis (TSA) have attracted significant attention. However, challenges such as training data scarcity and imbalance continue to hinder their development. Inspired by complex dynamic system theories, we design a series-symbol data generation mechanism, enabling the unrestricted creation of high-quality time series data paired with corresponding symbolic expressions. To leverage series-symbol data pairs with strong correlations, we develop SymTime, a pre-trained foundation model for enhancing time series representation using symbolic information. SymTime demonstrates competitive performance across five major TSA tasks when fine-tunes with downstream tasks, rivaling foundation models pre-trained on real-world datasets. This approach underscores the potential of series-symbol data generation and pretraining mechanisms in overcoming data scarcity and enhancing task performance. The code is available at https://github.com/wwhenxuan/SymTime.

cs.LG↗

arXiv2025

Deep Neural Networks for Modeling Astrophysical Nuclear Reacting Flows

In astrophysical simulations, nuclear reacting flows pose computational challenges due to the stiffness of reaction networks. We introduce neural network-based surrogate models using the DeePODE framework to enhance simulation efficiency while maintaining accuracy and robustness. Our method replaces conventional stiff ODE solvers with deep learning models trained through evolutionary Monte Carlo sampling from zero-dimensional simulation data, ensuring generalization across varied thermonuclear and hydrodynamic conditions. Tested on 3-species and 13-species reaction networks, the models achieve $\lesssim 1\%$ accuracy relative to semi-implicit numerical solutions and deliver a $\sim 2.6\times$ speedup on CPUs. A temperature-thresholded deployment strategy ensures stability in extreme conditions, sustaining neural network utilization above 75\% in multi-dimensional simulations. These data-driven surrogates effectively mitigate stiffness constraints, offering a scalable approach for high-fidelity modeling of astrophysical nuclear reacting flows.

astro-ph.IM↗

arXiv2025

Integrating Large Language Models and Reinforcement Learning for Sentiment-Driven Quantitative Trading

This research develops a sentiment-driven quantitative trading system that leverages a large language model, FinGPT, for sentiment analysis, and explores a novel method for signal integration using a reinforcement learning algorithm, Twin Delayed Deep Deterministic Policy Gradient (TD3). We compare the performance of strategies that integrate sentiment and technical signals using both a conventional rule-based approach and a reinforcement learning framework. The results suggest that sentiment signals generated by FinGPT offer value when combined with traditional technical indicators, and that reinforcement learning algorithm presents a promising approach for effectively integrating heterogeneous signals in dynamic trading environments.

q-fin.CP↗

arXiv2025

Exploring the Power of Diffusion Large Language Models for Software Engineering: An Empirical Investigation

Autoregressive Large Language Models (AR-LLMs) are widely used in software engineering (SE) but face limitations in processing code structure information and suffer from high inference latency. Diffusion LLMs (DLLMs) offer a promising alternative with global bidirectional encoding and decoupled generation steps. This work presents the first comprehensive evaluation of DLLMs across the software development lifecycle, including code generation, defect detection, and program repair. On a large-scale benchmark of 52,937 tasks, 7Bparameter DLLMs outperform AR-LLMs with a 30% average accuracy improvement achieving a 113% gain on cross-file repair, while maintaining superior efficiency and reduced latency. Our results establish DLLMs as a superior paradigm for SE tasks.

cs.SE↗

arXiv2025

AutoEmpirical: LLM-Based Automated Research for Empirical Software Fault Analysis

Understanding software faults is essential for empirical research in software development and maintenance. However, traditional fault analysis, while valuable, typically involves multiple expert-driven steps such as collecting potential faults, filtering, and manual investigation. These processes are both labor-intensive and time-consuming, creating bottlenecks that hinder large-scale fault studies in complex yet critical software systems and slow the pace of iterative empirical research. In this paper, we decompose the process of empirical software fault study into three key phases: (1) research objective definition, (2) data preparation, and (3) fault analysis, and we conduct an initial exploration study of applying Large Language Models (LLMs) for fault analysis of open-source software. Specifically, we perform the evaluation on 3,829 software faults drawn from a high-quality empirical study. Our results show that LLMs can substantially improve efficiency in fault analysis, with an average processing time of about two hours, compared to the weeks of manual effort typically required. We conclude by outlining a detailed research plan that highlights both the potential of LLMs for advancing empirical fault studies and the open challenges that required be addressed to achieve fully automated, end-to-end software fault analysis.

cs.SE↗

arXiv2025

Structural Heterogeneity of the Drosophila Brain Network

Decoding the heterogeneity of biological neural systems is key to understanding the nervous system's complex dynamical behaviors. This study analyzes the comprehensive Drosophila brain connectome, which is the most recent data set, containing over 130,000 neurons and 50 million synapses. We conducted meticulous analyses of both network and spatial structure. Our findings reveal significant heterogeneity in network properties and distinct spatial clustering across functional regions. Besides, our analysis revealed a modular organizational pattern within the neural network, wherein regions with similar functions exhibited higher connection densities, forming distinct community structures. Moreover, we observed spatial clustering within functional regions but was not statistically significant. Additionally, we identify pervasive bilateral symmetry in network topology and spatial organization. Simulations based on the Kuramoto model demonstrate that the functional asymmetry between cerebral hemispheres arises from disparities in the intrinsic frequencies of neurons rather than from structural asymmetry within the neural network itself. Finally, we develop a 3D connectome visualization tool for detailed mapping of neuronal morphology. These insights advance our understanding of neural network organization and complexity in biological systems.

q-bio.NC↗

arXiv2025

A Multidimensional Self-Adaptive Numerical Simulation Framework for Semiconductor Boltzmann Transport Equation

This research addresses the numerical simulation of the Boltzmann transport equation for semiconductor devices by proposing a multidimensional self-adaptive numerical simulation framework. This framework is applied to two important generalized forms of the equation: a parabolic equation with singular properties on the unit disk and a continuity equation. The study enhances the alignment of numerical simulations with physical characteristics through polar coordinate transformation and variable drift-diffusion coefficients. Innovatively, a multidimensional adaptive mesh partitioning strategy for radius-angle-time is designed and combined with an adjustable finite difference scheme to construct a highly adaptive numerical simulation method. In the construction of discrete schemes, the Swartztrauber-Sweet method and the control volume method are employed to effectively eliminate the origin singularity caused by polar coordinate transformation. On the programming front, a parallelized MATLAB algorithm is developed to optimize code execution efficiency. Numerical comparative experiments demonstrate that the adaptive method improves the accuracy of the parabolic equation by 1 to 7 times and that of the continuity equation by 10% to 70% while maintaining computational efficiency, significantly enhancing numerical simulation accuracy with high stability. Furthermore, this study systematically verifies the algorithm's convergence, stability, and parameter sensitivity using error visualization and other means. It also explores optimal parameters and establishes tuning optimization criteria. The research provides theoretical support for high-precision and highly adaptive methods in semiconductor device simulation, demonstrating outstanding advantages in handling singular regions.

math.NA↗