arXiv subjects

Xu Cai

Publications and source records attributed to Xu Cai.

At least 19 recordsLinked to original sources

arXiv2026

On the Convergence of the Current-Constrained Power Angle Curve of Virtual Admittance-Based Grid Forming Converters

The current-constrained power-angle curve (PAC) is crucial for the transient synchronization stability (TSS) analysis of virtual admittance-based (VA) grid-forming (GFM) converters. Its formulation and application rely on the quasi-steady-state assumption critically, i.e., the active power can converge to its steady-state across the entire angle space in both a stable and fast manner. Despite this assumption is intuitively perceivable, it lack sufficient clarification, particularly on underlying behaviors if violated. To this end, this paper uncovers a new phenomenon on the non-uniform convergence of the VA-PAC. To achieve this, an eigen-sweep-based analysis of the full-order VA-PAC model with detailed controls is conducted, by which the existence of this issue is theoretically confirmed. On this basis, the open-loop stability and response-rate conditions of the full-order VA-PAC model for ensuring its convergence are clarified. Findings of this work can provide deeper insights into the existing TSS analyses of GFM converters, and are expected to provoke new analyses.

eess.SY↗

arXiv2026

Generalized Feedback Control Modeling Method for Control-Driven Converter Systems

Converters-based systems like wind farms manifest themselves as control-intensive systems, where control-driven stability issues frequently occur, e.g., oscillations. Such issues are popularly studied via circuit impedance-based methods. However, given its implicit controller modeling trait, the impedance-based methods have limitations in system analysis and designs involving large-scale controllers. To address this issue, this paper presents a novel frequency domain modeling framework, as a perspective shift from the circuit to the control system. Since the obtained model features a multi-input-multi-output (MIMO) feedback control structure and explicit controller placement, it is termed the Generalized Feedback Control (GFC) model. GFC modeling is conducted for both single and multi-converter cases, and the resulting models are validated by frequency scan and stability test. Moreover, advantages of the GFC method in achieving interaction analysis and stability-oriented designs of multi-controllers are demonstrated by three application examples, further suggesting its great potential for being applied to the analysis and design issues of converter systems involving large-scale controllers.

eess.SY↗

arXiv2025

DTRec: Learning Dynamic Reasoning Trajectories for Sequential Recommendation

Inspired by advances in LLMs, reasoning-enhanced sequential recommendation performs multi-step deliberation before making final predictions, unlocking greater potential for capturing user preferences. However, current methods are constrained by static reasoning trajectories that are ill-suited for the diverse complexity of user behaviors. They suffer from two key limitations: (1) a static reasoning direction, which uses flat supervision signals misaligned with human-like hierarchical reasoning, and (2) a fixed reasoning depth, which inefficiently applies the same computational effort to all users, regardless of pattern complexity. These rigidity lead to suboptimal performance and significant computational waste. To overcome these challenges, we propose DTRec, a novel and effective framework that explores the Dynamic reasoning Trajectory for Sequential Recommendation along both direction and depth. To guide the direction, we develop Hierarchical Process Supervision (HPS), which provides coarse-to-fine supervisory signals to emulate the natural, progressive refinement of human cognitive processes. To optimize the depth, we introduce the Adaptive Reasoning Halting (ARH) mechanism that dynamically adjusts the number of reasoning steps by jointly monitoring three indicators. Extensive experiments on three real-world datasets demonstrate the superiority of our approach, achieving up to a 24.5% performance improvement over strong baselines while simultaneously reducing computational cost by up to 41.6%.

cs.IR↗

arXiv2025

Shortcutting Pre-trained Flow Matching Diffusion Models is Almost Free Lunch

We present an ultra-efficient post-training method for shortcutting large-scale pre-trained flow matching diffusion models into efficient few-step samplers, enabled by novel velocity field self-distillation. While shortcutting in flow matching, originally introduced by shortcut models, offers flexible trajectory-skipping capabilities, it requires a specialized step-size embedding incompatible with existing models unless retraining from scratch$\unicode{x2013}$a process nearly as costly as pretraining itself. Our key contribution is thus imparting a more aggressive shortcut mechanism to standard flow matching models (e.g., Flux), leveraging a unique distillation principle that obviates the need for step-size embedding. Working on the velocity field rather than sample space and learning rapidly from self-guided distillation in an online manner, our approach trains efficiently, e.g., producing a 3-step Flux less than one A100 day. Beyond distillation, our method can be incorporated into the pretraining stage itself, yielding models that inherently learn efficient, few-step flows without compromising quality. This capability also enables, to our knowledge, the first few-shot distillation method (e.g., 10 text-image pairs) for dozen-billion-parameter diffusion models, delivering state-of-the-art performance at almost free cost.

cs.CV↗

arXiv2025

FlexControl: Computation-Aware ControlNet with Differentiable Router for Text-to-Image Generation

ControlNet offers a powerful way to guide diffusion-based generative models, yet most implementations rely on ad-hoc heuristics to choose which network blocks to control-an approach that varies unpredictably with different tasks. To address this gap, we propose FlexControl, a novel framework that copies all diffusion blocks during training and employs a trainable gating mechanism to dynamically select which blocks to activate at each denoising step. With introducing a computation-aware loss, we can encourage control blocks only to activate when it benefit the generation quality. By eliminating manual block selection, FlexControl enhances adaptability across diverse tasks and streamlines the design pipeline, with computation-aware training loss in an end-to-end training manner. Through comprehensive experiments on both UNet (e.g., SD1.5) and DiT (e.g., SD3.0), we show that our method outperforms existing ControlNet variants in certain key aspects of interest. As evidenced by both quantitative and qualitative evaluations, FlexControl preserves or enhances image fidelity while also reducing computational overhead by selectively activating the most relevant blocks. These results underscore the potential of a flexible, data-driven approach for controlled diffusion and open new avenues for efficient generative model design. The code will soon be available at https://github.com/Anonymousuuser/FlexControl.

cs.LG↗

arXiv2025

TSMini: A Simple Yet Highly Effective Trajectory Similarity Learning Model

Trajectory similarity is fundamental to many spatio-temporal data mining applications. Recent studies propose deep learning models to approximate conventional trajectory similarity measures, exploiting their fast inference time once trained. Although efficient inference has been reported, challenges remain in similarity approximation accuracy due to difficulties in trajectory granularity modeling and in exploiting similarity signals in training data. To fill this gap, we propose TSMini, a highly effective trajectory similarity model with a sub-view modeling mechanism and a k nearest neighbor-based loss. The former enables learning multi-granularity trajectory patterns, while the latter guides TSMini to learn not only absolute similarity values between trajectories but also their relative similarity ranks. Together, these innovations enable highly accurate trajectory similarity approximation. Experiments show that TSMini outperforms the state-of-the-art models by 15% on average when learning widely used trajectory similarity measures.

cs.LG↗

arXiv2024

No More Adam: Learning Rate Scaling at Initialization is All You Need

In this work, we question the necessity of adaptive gradient methods for training deep neural networks. SGD-SaI is a simple yet effective enhancement to stochastic gradient descent with momentum (SGDM). SGD-SaI performs learning rate Scaling at Initialization (SaI) to distinct parameter groups, guided by their respective gradient signal-to-noise ratios (g-SNR). By adjusting learning rates without relying on adaptive second-order momentum, SGD-SaI helps prevent training imbalances from the very first iteration and cuts the optimizer's memory usage by half compared to AdamW. Despite its simplicity and efficiency, SGD-SaI consistently matches or outperforms AdamW in training a variety of Transformer-based tasks, effectively overcoming a long-standing challenge of using SGD for training Transformers. SGD-SaI excels in ImageNet-1K classification with Vision Transformers(ViT) and GPT-2 pretraining for large language models (LLMs, transformer decoder-only), demonstrating robustness to hyperparameter variations and practicality for diverse applications. We further tested its robustness on tasks like LoRA fine-tuning for LLMs and diffusion models, where it consistently outperforms state-of-the-art optimizers. From a memory efficiency perspective, SGD-SaI achieves substantial memory savings for optimizer states, reducing memory usage by 5.93 GB for GPT-2 (1.5B parameters) and 25.15 GB for Llama2-7B compared to AdamW in full-precision training settings.

cs.LG↗

arXiv2024

Lower Bounds for Time-Varying Kernelized Bandits

The optimization of black-box functions with noisy observations is a fundamental problem with widespread applications, and has been widely studied under the assumption that the function lies in a reproducing kernel Hilbert space (RKHS). This problem has been studied extensively in the stationary setting, and near-optimal regret bounds are known via developments in both upper and lower bounds. In this paper, we consider non-stationary scenarios, which are crucial for certain applications but are currently less well-understood. Specifically, we provide the first algorithm-independent lower bounds, where the time variations are subject satisfying a total variation budget according to some function norm. Under $\ell_{\infty}$-norm variations, our bounds are found to be close to an existing upper bound (Hong et al., 2023). Under RKHS norm variations, the upper and lower bounds are still reasonably close but with more of a gap, raising the interesting open question of whether non-minor improvements in the upper bound are possible.

stat.ML↗

arXiv2024

Frequency-Reactive Power Optimization Strategy of Grid-forming Offshore Wind Farm Using DRU-HVDC Transmission

The diode rectifier unit-based high voltage direct current (DRU-HVDC) transmission with grid-forming (GFM) wind turbine is becoming a promising scheme for offshore wind farm(OWF) integration due to its high reliability and low cost. In this scheme, the AC network of the OWF and the DRU has completely different synchronization mechanisms and power flow characteristics from the traditional power system. To optimize the power flow and reduce the net loss, this paper carries out the power flow modeling and optimization analysis for the DRU-HVDC transmission system with grid-forming OWFs. The influence of the DRU and the GFM wind turbines on the power flow of the system is analyzed. On this basis, improved constraint conditions are proposed and an optimal power flow (OPF) method is established. This method can minimize the power loss by adjusting the reactive power output of each wind turbine and internal network frequency. Finally, based on MATLAB, this paper uses YALMIP toolkit and CPLEX mathematical solver to realize the programming solution of the OPF model proposed in this paper. The results show that the proposed optimization strategy can effectively reduce the power loss of the entire OWF and the transmission system with an optimization ratio of network losses exceeding 25.3%.

math.OC↗

arXiv2024

A Multilayer Eigen-Sensitivity Method Using Loop Gain Model for Oscillation Diagnosis of Converter-Based System

Loop gain-based eigen-sensitivity (LGES) is a useful frequency-domain tool for oscillation diagnosis of converter-based system. However, the existing theory is still scant in two aspects: participation factor (PF) is bound up with the frequency-domain modal characteristic that does not necessarily point to the stability as that of the time-domain eigen-sensitivity (i.e., PF of oscillation mode); a systematic LGES analysis framework containing both component- and parameter- level sensitivity is missing. These two factors hinder the application of LGES method on the proper evaluation of stability effects, which are closely related with the time-domain oscillation mode. To address these issues, this paper proposes a multilayer LGES method directed to the oscillation mode, and a full set of indices like PF, component and parameter sensitivity are established. The link from the eigen-sensitivity of frequency domain to that of time domain is revealed, through which it is shown how the proposed LGES method can facilitate the control parameter tuning-guided oscillation suppression. The effectiveness of the proposed LGES method is validated via case studies conducted on a generic AC/DC converter-based system.

physics.app-ph↗

arXiv2024

Kernelized Normalizing Constant Estimation: Bridging Bayesian Quadrature and Bayesian Optimization

In this paper, we study the problem of estimating the normalizing constant $\int e^{-\lambda f(x)}dx$ through queries to the black-box function $f$, where $f$ belongs to a reproducing kernel Hilbert space (RKHS), and $\lambda$ is a problem parameter. We show that to estimate the normalizing constant within a small relative error, the level of difficulty depends on the value of $\lambda$: When $\lambda$ approaches zero, the problem is similar to Bayesian quadrature (BQ), while when $\lambda$ approaches infinity, the problem is similar to Bayesian optimization (BO). More generally, the problem varies between BQ and BO. We find that this pattern holds true even when the function evaluations are noisy, bringing new aspects to this topic. Our findings are supported by both algorithm-independent lower bounds and algorithmic upper bounds, as well as simulation studies conducted on a variety of benchmark functions.

cs.LG↗

arXiv2023

An introduction to the parton and hadron cascade model PACIAE 3.0

We introduce a parton and hadron cascade model PACIAE 3.0 based on PYTHIA 6.428 and the PACIAE 2.2 program series. The simulation framework of C-, B-, and A-loops are designed for the high energy ($\sqrt{s_{NN}}\geq 3$ GeV) and low energy ($\sqrt{s_{NN}}<3$ GeV) nuclear collisions, respectively, in PACIAE 3.0. In the C-loop simulation, the parton-parton inelastic scattering processes are added in the partonic rescattering process. The single string structure and multiple string interaction mechanism have been introduced investigating the strangeness enhancement in C- and B-loop. An improved mapping relation between the centrality percentage definition and the impact parameter definition is proposed responding the observation of $b_{max}\approx 20$ fm from ALICE, ATLAS, and CMS collaborations. We have extensively modified the phenomenological coalescence hadronization model. The PACIAE 3.0 model simulated results of particle yield, transverse momentum distribution, and rapidity distribution well reproduce, respectively, the experimental data measured at FOPI, E895, RHIC, and LHC energies.

hep-ph↗

arXiv2022

Revisiting the centrality definition and observable centrality dependence of relativistic heavy-ion collisions in PACIAE model

We improve the centrality definition in impact parameter in PACIAE model responding the fact reported by the ALICE, ATLAS, and CMS collaborations that the maximum impact parameter in heavy ion collisions should be extended to 20 $fm$. Meanwhile the PACIAE program is updated to a new version of PACIAE 2.2.2 with convenience of studying the elementary nuclear collisions, proton-nucleus collisions, and the nucleus-nucleus collisions in one unified program version. The new impact parameter definition together with the optical Glauber model calculated impact parameter bin, $N_{part}$, and $N_{coll}$ in proton-nucleus and nucleus-nucleus collisions at relativistic energies are consistent with the improved MC-Glauber model ones within the error bar. The charged-particle pseudorapidity and the transverse momentum distributions in Pb-Pb collisions at $\sqrt{s_{\rm NN}}=5.02$ TeV simulated by PACIAE 2.2.2 well reproduce the ALICE experimental data.

nucl-th↗

arXiv2022

Linear scaling of lepton charge asymmetry in $W^\pm$ production in ultra-relativistic nuclear collisions

The lepton charge asymmetry in $W^\pm$ production in the nuclear collisions at $\sqrt{s_{\rm NN}}=5.02$ TeV is investigated with a parton and hadron cascade model PACIAE. Recently published ALICE and the ATLAS data of lepton charge asymmetry are well reproduced. An interesting linear scaling behavior is observed in the lepton charge asymmetry as a function of the collision system valence quark number asymmetry among the different size of nuclear collision systems at $\sqrt{s_{\rm NN}}=5.02$ TeV. This linear scaling behavior may serve as an additional constraint on the PDF (nPDF) extractions.

hep-ph↗

arXiv2022

Systematic study of hadrons and their quark-component nuclear modification factors

We have systematically studied the connection between hadron and its quark component nuclear modification factors and the flavor (mass) ordering at both parton and hadron levels in the nucleus-nucleus collisions at the LHC energies by the PACIAE model. It turns out that these two physical phenomena explored in our last publication (J. Phys. G 49 065104, 2022) are generally held, irrespective of the rapidity, centrality, reaction energy, and the collision system size. The mass ordering at hadron level in nuclear modification factor seems to be really existed, which should be studied further both theoretically and experimentally.

hep-ph↗

arXiv2022

Three-Port Impedance Model and Validation of VSCs for Stability Analysis

Modern power system is undergoing a paradigm shift from the synchronous generators-based system to the power electronics converters-dominated system. With the high penetration of converters, serious stability problems are provoked, especially the wideband oscillations. Various studies have been conducted in this respect, while most of them separate the ac-side stability with the dc-side stability. However, for the stability analysis of the hybrid AC/DC grid, it is necessary to consider the converter ac-side and dc-side, simultaneously. In this paper, the stability analysis of voltage source converters (VSCs) considering both ac and dc dynamics is carried out. At first, the three-port AC/DC admittance model of VSCs is established, and the corresponding measurement method from simulations is presented to validate its accuracy. Secondly, based on such three-port model, two stability analysis methods are presented: the one is based on the system open-loop model, where the stability can be judged via the Generalized Nyquist Criterion (GNC); the other one is based on the system closed-loop model, whose stability can be predicted through the pole-zero calculation. At last, a test AC/DC system is built in MATLAB/Simulink, by which the effectiveness of the three-port model-based stability analysis is validated.

physics.app-ph↗

arXiv2022

On Average-Case Error Bounds for Kernel-Based Bayesian Quadrature

In this paper, we study error bounds for {\em Bayesian quadrature} (BQ), with an emphasis on noisy settings, randomized algorithms, and average-case performance measures. We seek to approximate the integral of functions in a {\em Reproducing Kernel Hilbert Space} (RKHS), particularly focusing on the Mat\'ern-$\nu$ and squared exponential (SE) kernels, with samples from the function potentially being corrupted by Gaussian noise. We provide a two-step meta-algorithm that serves as a general tool for relating the average-case quadrature error with the $L^2$-function approximation error. When specialized to the Mat\'ern kernel, we recover an existing near-optimal error rate while avoiding the existing method of repeatedly sampling points. When specialized to other settings, we obtain new average-case results for settings including the SE kernel with noise and the Mat\'ern kernel with misspecification. Finally, we present algorithm-independent lower bounds that have greater generality and/or give distinct proofs compared to existing ones.

stat.ML↗

arXiv2022

Centrality dependence and isospin effect on $\Wpm$ and $\Zn$ productions in nucleus-nucleus collisions at $\sNN=5.02$~TeV

In this paper, the centrality dependent $\Zn$ and $\Wpm$ production and the isospin effect in $\Wpm$ production are investigated with a parton and hadron cascade model PACIAE in Pb--Pb collisions at $\sNN=5.02$~TeV. ALICE data of $\Zn$ production in Pb--Pb collisions at $\sNN=5.02$~TeV are found to be reproduced fairly well. The prediction on $\Wpm$ production in the same collision system is given as well. An interesting isospin effect is observed in exploring the charge asymmetry between $\Wp$ and $\Wm$ as a function of the asymmetry between number of valence $u$- and $d$-quarks varied from small to large collision systems at center-of-mass energy 5.02 TeV. The results serve as a important benchmarks for understanding the initial conditions of heavy-ion collisions.

hep-ph↗