arXiv subjects

Peng Chen

Publications and source records attributed to Peng Chen.

At least 181 records · Page 10Linked to original sources

arXiv2025

A Dynamic and High-Precision Method for Scenario-Based HRA Synthetic Data Collection in Multi-Agent Collaborative Environments Driven by LLMs

HRA (Human Reliability Analysis) data is crucial for advancing HRA methodologies. however, existing data collection methods lack the necessary granularity, and most approaches fail to capture dynamic features. Additionally, many methods require expert knowledge as input, making them time-consuming and labor-intensive. To address these challenges, we propose a new paradigm for the automated collection of HRA data. Our approach focuses on key indicators behind human error, specifically measuring workload in collaborative settings. This study introduces a novel, scenario-driven method for workload estimation, leveraging fine-tuned large language models (LLMs). By training LLMs on real-world operational data from high-temperature gas-cooled reactors (HTGRs), we simulate human behavior and cognitive load in real time across various collaborative scenarios. The method dynamically adapts to changes in operator workload, providing more accurate, flexible, and scalable workload estimates. The results demonstrate that the proposed WELLA (Workload Estimation with LLMs and Agents) outperforms existing commercial LLM-based methods in terms of prediction accuracy.

cs.AI↗

arXiv2025

UIFV: Data Reconstruction Attack in Vertical Federated Learning

Vertical Federated Learning (VFL) facilitates collaborative machine learning without the need for participants to share raw private data. However, recent studies have revealed privacy risks where adversaries might reconstruct sensitive features through data leakage during the learning process. Although data reconstruction methods based on gradient or model information are somewhat effective, they reveal limitations in VFL application scenarios. This is because these traditional methods heavily rely on specific model structures and/or have strict limitations on application scenarios. To address this, our study introduces the Unified InverNet Framework into VFL, which yields a novel and flexible approach (dubbed UIFV) that leverages intermediate feature data to reconstruct original data, instead of relying on gradients or model details. The intermediate feature data is the feature exchanged by different participants during the inference phase of VFL. Experiments on four datasets demonstrate that our methods significantly outperform state-of-the-art techniques in attack precision. Our work exposes severe privacy vulnerabilities within VFL systems that pose real threats to practical VFL applications and thus confirms the necessity of further enhancing privacy protection in the VFL architecture.

cs.LG↗

arXiv2025

Giant and Flexible Toroidal Circular Dichroism from Planar Chiral Metasurface

Chirality, a fundamental concept describing an object cannot superpose with its mirror image, is crucial in optics and photonics and leads to various exotic phenomena, such as circular dichroism, and optical activity. Recent findings reveal that, besides electric and magnetic dipoles, toroidal dipoles, an elusive part of dynamic multipoles, can also contribute significantly to chirality. However, as toroidal dipoles are typically represented by solenoidal currents circulating on a three-dimensional (3D) torus, toroidal circular dichroism is usually observed in 3D intricate microstructures. Facing corresponding challenges in fabrication, integration and application, it is generally difficult to employ toroidal circular dichroism in compact metasurfaces for flexible modulation of chiral interactions between electromagnetic waves and matter. To overcome these stringent challenges, we propose and experimentally demonstrate the giant toroidal circular dichroism in a bilayer metasurface that is comprised of only planar layers, effectively bypassing various restrictions imposed by 3D microstructures. With the introduction of a displacement, or bilayer offset, between the opposite layers, we experimentally achieve giant chiral responses with the intrinsic circular dichroism (CD) reaching 0.69 in measurements, and the CD can be quantitatively manipulated in a simple manner. The giant intrinsic chirality primarily originates from distinct excitations of in-plane toroidal dipole moments under circular polarized incidences, and the toroidal chiral response is quantitatively controlled by the bilayer offset. Therefore, our work provides a straightforward and versatile approach for development of giant and flexible intrinsic chirality through toroidal dipoles with inherently planar layers, important for applications in communications, sensing, and chiroptical devices.

physics.optics↗

arXiv2025

Qubit-assisted quantum metrology under a time-reversal strategy

We propose a quantum metrology protocol based on a two-step joint evolution of the probe system and an ancillary qubit and quantum measurement. With a proper initial state of the ancillary qubit and an optimized evolution time, the quantum Fisher information (QFI) about the phase parameter encoded in the probe system is found to be determined by the expectation value of the square of a phase generator, irrespective of the probe initial state. Consequently, even if the probe is prepared as a finite-temperature state, faraway from the so-called resource state, e.g., the squeezed spin state or the Greenberger-Horne-Zeilinger state in atomic systems, the QFI in our protocol can approach the Heisenberg scaling $N^2$ with respect to the probe size $N$. This quadratic scaling behavior shows robustness against the imperfections about the initial state of the ancillary qubit and the optimized evolution time of the whole system. Also it is not sensitive to the deviation in system parameters and the qubit decoherence. Using the time-reversal strategy and a single-shot projective measurement, the classical Fisher information (CFI) in our metrology protocol is saturated with its quantum counterpart. Our work thus provides an economical method to reach the Heisenberg limit in metrology precision with no input of entanglement or squeezing.

quant-ph↗

arXiv2024

Study on the efficiency droop in high-quality GaN material under high photoexcitation intensity

III-V nitride semiconductors, represented by GaN, have attracted significant research attention. Driven by the growing interest in smart micro-displays, there is a strong desire to achieve enhanced light output from even smaller light-emitting diode (LED) chips. However, the most perplexing phenomenon and the most significant challenge in the study of emission properties under high-injection conditions in GaN has always been efficiency droop for decades, where LEDs exhibit a substantial loss in efficiency at high driving currents. In this paper, we present our study on the intrinsic emission properties of high-quality GaN material based on the density of states and the principles of momentum conservation. Our theoretical calculations reveal a momentum distribution mismatch between the non-equilibrium excess electrons and holes, which becomes more significant as the carrier concentration increases. Our excitation-dependent photoluminescence measurements conducted at 6 K exhibited a clear droop for all exciton recombinations, but droop-free for phonon-assisted recombination due to phonons compensating for the momentum mismatch. These findings indicate that the momentum distribution mismatch between the non-equilibrium excess electrons and holes is one of the intrinsic causes of the efficiency droop, which originates from the intrinsic band properties of GaN. These results suggest that proper active region design aimed at reducing this mismatch will contribute to the development of ultra-highly efficient lighting devices in the future.

physics.optics↗

arXiv2024

Enhanced optical performance of GaN Micro-light-emitting diodes with a single porous layer

High-efficiency micro-light-emitting diodes (Micro-LEDs) are key devices for next-generation display technology. However, when the mesa size is reduced to around tens of micrometers or less, the luminous efficiency is constrained by the "efficiency-on-size effect". This work details the fabrication of gallium nitride (GaN) based Micro-LEDs with various mesa shapes and a single porous layer under the active region. A modified green LED epitaxial structure with different doped n-GaN layers combined with electrochemical etching created the porous layer. The strong light confinement achieved by the porous layer and the polygonal mesa greatly enhances spontaneous emission. The luminous intensity of the Micro-LEDs with the porous layer is approximately 22 times greater than those Micro-LEDs without the porous layer. A significant reduction in minimum full width at half maximum (FWHM) was observed in polygonal devices, suggesting a change in the luminescence mechanism. The influence of varying device geometry on emission performance was investigated. Experimental results reveal that, unlike circular porous Micro-LEDs, square and hexagonal porous Micro-LEDs exhibit more pronounced resonant emission, which provides a new technological approach for the further development of high-performance Micro-LEDs and lasers.

physics.optics↗

arXiv2024

Cavity Plasmon: Enhanced Luminescence Effect on InGaN Light Emitting Diodes

We fabricated polygonal nanoholes in the top p-GaN layer of the InGaN/GaN light-emitting diode, followed by the deposition of Au/Al metal thin film within the nanoholes to create metal microcavities, thereby constructing the surface plasmon structure. The findings indicate that with increased current injection, the light output of the LEDs rose by 46%, accompanied by a shift of the gain peak position towards the plasmon resonance energy. The maximum enhancement factor increases to 2.38 as the coupling distance decreases from 60 nm to 30 nm. Interestingly, time-resolved photoluminescence data showed that the spontaneous emission decay time lengthened due to the plasmon coupling, suggesting the presence of a new plasmon coupling mechanism. Finite-Difference Time-Domain simulation results show that the electric field is localized at certain locations around the metal microcavity, generating a new type of shape-sensitive plasmon, named Cavity Plasmon here. This intense localization leads to a longer lifetime and enhances the recombination efficiency of excitons. We discuss several unique properties of the cavity plasmon generated by the polygonal metal microcavity with several specific angular shapes. The results demonstrate that the cavity plasmon generated by the polygonal metal microcavity is a highly promising technique for enhancing the light emission performance of of relevant semiconductor optoelectronic devices.

physics.optics↗

arXiv2024

A Novel Task-Driven Method with Evolvable Interactive Agents Using Event Trees for Enhanced Emergency Decision Support

As climate change and other global challenges increase the likelihood of unforeseen emergencies, the limitations of human-driven strategies in critical situations become more pronounced. Inadequate pre-established emergency plans can lead operators to become overwhelmed during complex systems malfunctions. This study addresses the urgent need for agile decision-making in response to various unforeseen incidents through a novel approach, EvoTaskTree (a task-driven method with evolvable interactive agents using event trees for emergency decision support). This advanced approach integrates two types of agents powered by large language models (LLMs): task executors, responsible for executing critical procedures, and task validators, ensuring the efficacy of those actions. By leveraging insights from event tree analysis, our framework encompasses three crucial tasks: initiating event subevent analysis, event tree header event analysis, and decision recommendations. The agents learn from both successful and unsuccessful responses from these tasks. Finally, we use nuclear power plants as a demonstration of a safety-critical system. Our findings indicate that the designed agents are not only effective but also outperform existing approaches, achieving an impressive accuracy rate of up to 100 % in processing previously unencoun32 tered incident scenarios. This paper demonstrates that EvoTaskTree significantly enhances the rapid formulation of emergency decision-making.

cs.AI↗

arXiv2024

KRAIL: A Knowledge-Driven Framework for Base Human Reliability Analysis Integrating IDHEAS and Large Language Models

Human reliability analysis (HRA) is crucial for evaluating and improving the safety of complex systems. Recent efforts have focused on estimating human error probability (HEP), but existing methods often rely heavily on expert knowledge,which can be subjective and time-consuming. Inspired by the success of large language models (LLMs) in natural language processing, this paper introduces a novel two-stage framework for knowledge-driven reliability analysis, integrating IDHEAS and LLMs (KRAIL). This innovative framework enables the semi-automated computation of base HEP values. Additionally, knowledge graphs are utilized as a form of retrieval-augmented generation (RAG) for enhancing the framework' s capability to retrieve and process relevant data efficiently. Experiments are systematically conducted and evaluated on authoritative datasets of human reliability. The experimental results of the proposed methodology demonstrate its superior performance on base HEP estimation under partial information for reliability assessment.

cs.CL↗

arXiv2024

GraphAvatar: Compact Head Avatars with GNN-Generated 3D Gaussians

Rendering photorealistic head avatars from arbitrary viewpoints is crucial for various applications like virtual reality. Although previous methods based on Neural Radiance Fields (NeRF) can achieve impressive results, they lack fidelity and efficiency. Recent methods using 3D Gaussian Splatting (3DGS) have improved rendering quality and real-time performance but still require significant storage overhead. In this paper, we introduce a method called GraphAvatar that utilizes Graph Neural Networks (GNN) to generate 3D Gaussians for the head avatar. Specifically, GraphAvatar trains a geometric GNN and an appearance GNN to generate the attributes of the 3D Gaussians from the tracked mesh. Therefore, our method can store the GNN models instead of the 3D Gaussians, significantly reducing the storage overhead to just 10MB. To reduce the impact of face-tracking errors, we also present a novel graph-guided optimization module to refine face-tracking parameters during training. Finally, we introduce a 3D-aware enhancer for post-processing to enhance the rendering quality. We conduct comprehensive experiments to demonstrate the advantages of GraphAvatar, surpassing existing methods in visual fidelity and storage consumption. The ablation study sheds light on the trade-offs between rendering quality and model size. The code will be released at: https://github.com/ucwxb/GraphAvatar

cs.CV↗

arXiv2024

A Hybrid Real-Time Framework for Efficient Fussell-Vesely Importance Evaluation Using Virtual Fault Trees and Graph Neural Networks

The Fussell-Vesely Importance (FV) reflects the potential impact of a basic event on system failure, and is crucial for ensuring system reliability. However, traditional methods for calculating FV importance are complex and time-consuming, requiring the construction of fault trees and the calculation of minimal cut set. To address these limitations, this study proposes a hybrid real-time framework to evaluate the FV importance of basic events. Our framework combines expert knowledge with a data-driven model. First, we use Interpretive Structural Modeling (ISM) to build a virtual fault tree that captures the relationships between basic events. Unlike traditional fault trees, which include intermediate events, our virtual fault tree consists solely of basic events, reducing its complexity and space requirements. Additionally, our virtual fault tree considers the dependencies between basic events rather than assuming their independence, as is typically done in traditional fault trees. We then feed both the event relationships and relevant data into a graph neural network (GNN). This approach enables a rapid, data-driven calculation of FV importance, significantly reducing processing time and quickly identifying critical events, thus providing robust decision support for risk control. Results demonstrate that our model performs well in terms of MSE, RMSE, MAE, and R2, reducing computational energy consumption and offering real-time, risk-informed decision support for complex systems.

cs.LG↗

arXiv2024

Stable Approximation for Call Function Via Stein's method

Let $S_{n}$ be a sum of independent identically distribution random variables with finite first moment and $h_{M}$ be a call function defined by $g_{M}(x)=\max\{x-M,0\}$ for $x\in\mathbb{R}$, $M>0$. In this paper, we assume the random variables are in the domain $\mathcal{R}_α$ of normal attraction of a stable law of exponent $α$, then for $α\in(1,2)$, we use the Stein's method developed in \cite{CNX21} to give uniform and non uniform bounds on $α$-stable approximation for the call function without additional moment assumptions. These results will make the approximation theory of call function applicable to the lower moment conditions, and greatly expand the scope of application of call function in many fields.

math.PR↗

arXiv2024

$W_{\bf d}$-convergence rate of EM schemes for invariant measures of supercritical stable SDEs

By establishing the regularity estimates for nonlocal Stein/Poisson equations under $γ$-order Hölder and dissipative conditions on the coefficients, we derive the $W_{\bf d}$-convergence rate for the Euler-Maruyama schemes applied to the invariant measure of SDEs driven by multiplicative $α$-stable noises with $α\in (\frac{1}{2}, 2)$, where $W_{\bf d}$ denotes the Wasserstein metric with ${\bf d}(x,y)=|x-y|^γ\wedge 1$ and $γ\in ((1-α)_+, 1]$.

math.PR↗

arXiv2024

A Link-Based Flow Model with Turn-Level Queue Transmission and Time-Varying Free-Flow Speed for Urban Road Networks

Macroscopic link-based flow models are efficient for simulating flow propagation in urban road networks. Existing link-based flow models described traffic states of a link with two state variables of link inflow and outflow and assumed homogeneous traffic states within a whole link. Consequently, the turn-level queue length change within the link can not be captured, resulting in underrepresented queue spillback. Moreover, a constant link free-flow speed was assumed to formulate models, restricting their applicability in modeling phenomena involving time-varying free-flow speed. This study proposed a new link-based flow model by introducing an additional state variable of link queue inflow and adapting the link outflow to be free-flow speed-dependent. In our model, the vehicle propagation within each link is described by the link inflow, queue inflow, and outflow, which depends on the link free-flow speed changes. A node model is further defined to capture the presence of signal control and potential queue spillback, which estimates the constrained flow propagation between adjacent road segments. Simulation experiments were conducted on a single intersection and a network with consecutive intersections to verify the proposed model performance. Results demonstrate the predictive power of the proposed model in predicting traffic operations of intersections with multiple turning movements and time-varying free-flow speed. Our model outperforms the baseline link-based flow model and preserves the computational tractability property of link-based flow models.

eess.SY↗

arXiv2024

Secure Antenna Selection and Beamforming in MIMO Systems

This work proposes a novel joint design for multiuser multiple-input multiple-output wiretap channels. The base station exploits a switching network to connect a subset of its antennas to the available radio frequency chains. The switching network and transmit beamformers are jointly designed to maximize the weighted secrecy sum-rate for this setting. The principal design problem reduces to an NP-hard mixed-integer non-linear programming. We invoke the fractional programming technique and the penalty dual decomposition method to develop a tractable iterative algorithm that effectively approximates the optimal design. Our numerical investigations validate the effectiveness of the proposed algorithm and its superior performance compared with the benchmark.

eess.SP↗

arXiv2024

Derivative-enhanced Deep Operator Network

The deep operator networks (DeepONet), a class of neural operators that learn mappings between function spaces, have recently been developed as surrogate models for parametric partial differential equations (PDEs). In this work we propose a derivative-enhanced deep operator network (DE-DeepONet), which leverages derivative information to enhance the solution prediction accuracy and provides a more accurate approximation of solution-to-parameter derivatives, especially when training data are limited. DE-DeepONet explicitly incorporates linear dimension reduction of high dimensional parameter input into DeepONet to reduce training cost and adds derivative loss in the loss function to reduce the number of required parameter-solution pairs. We further demonstrate that the use of derivative loss can be extended to enhance other neural operators, such as the Fourier neural operator (FNO). Numerical experiments validate the effectiveness of our approach.

cs.LG↗

arXiv2024

PlantCamo: Plant Camouflage Detection

Camouflaged Object Detection (COD) aims to detect objects with camouflaged properties. Although previous studies have focused on natural (animals and insects) and unnatural (artistic and synthetic) camouflage detection, plant camouflage has been neglected. However, plant camouflage plays a vital role in natural camouflage. Therefore, this paper introduces a new challenging problem of Plant Camouflage Detection (PCD). To address this problem, we introduce the PlantCamo dataset, which comprises 1,250 images with camouflaged plants representing 58 object categories in various natural scenes. To investigate the current status of plant camouflage detection, we conduct a large-scale benchmark study using 20+ cutting-edge COD models on the proposed dataset. Due to the unique characteristics of plant camouflage, including holes and irregular borders, we developed a new framework, named PCNet, dedicated to PCD. Our PCNet surpasses performance thanks to its multi-scale global feature enhancement and refinement. Finally, we discuss the potential applications and insights, hoping this work fills the gap in fine-grained COD research and facilitates further intelligent ecology research. All resources will be available on https://github.com/yjybuaa/PlantCamo.

cs.CV↗

arXiv2024

Learning-Augmented Algorithms for the Bahncard Problem

In this paper, we study learning-augmented algorithms for the Bahncard problem. The Bahncard problem is a generalization of the ski-rental problem, where a traveler needs to irrevocably and repeatedly decide between a cheap short-term solution and an expensive long-term one with an unknown future. Even though the problem is canonical, only a primal-dual-based learning-augmented algorithm was explicitly designed for it. We develop a new learning-augmented algorithm, named PFSUM, that incorporates both history and short-term future to improve online decision making. We derive the competitive ratio of PFSUM as a function of the prediction error and conduct extensive experiments to show that PFSUM outperforms the primal-dual-based algorithm.

cs.LG↗