arXiv subjects

Liang Xue

Publications and source records attributed to Liang Xue.

18 recordsLinked to original sources

arXiv2026

Flow Reversal in Low-Prandtl-Number Convection via Lateral Confinement

A prevailing consensus holds that flow reversals of the large-scale circulation (LSC) are suppressed in low-Prandtl-number (Pr) fluids, as high thermal diffusivity rapidly dissipates the energy required to fuel the corner-vortex mechanisms. Here, we report Direct Numerical Simulations of liquid metal convection (Pr=0.029) revealing that strong lateral confinement defies this consensus, enabling sustained LSC reversals. We show that confinement triggers a ``plume condensation" transition, reorganizing chaotic thermal plumes into highly coherent, quasi-linear structures. A thermal dissipation analysis demonstrates that this coherence drastically reduces heat loss during transport, allowing plumes to deliver sufficient buoyancy to corner vortices to drive reversals. We map a distinct ``island of reversal" in the parameter space, establishing lateral confinement as a control parameter capable of overcoming the stabilizing effects of high thermal diffusivity.

physics.flu-dyn↗

arXiv2026

STREAM: A Data-Centric Framework for Mining High-Value Task-Oriented Dialogues from Streaming Media

Large language models for vertical domains are bottlenecked by the scarcity of complex, domain-specific task-oriented dialogues. Existing data acquisition pipelines face a persistent trilemma: expert annotation is expensive, real-world service conversations are constrained by privacy and commercial restrictions, and static corpora quickly become temporally stale. We propose Stream, a data-centric framework that leverages publicly available streaming media (live streams and short videos) to synthesize high-value service dialogues at scale. Stream mines authentic interaction signals from noisy streams and synthesizes conversations by integrating role-grounded persona construction with Conversational Blueprint construction; it further adopts retrieval-augmented generation (RAG) to support knowledge-aware responses. Based on Stream, we release StreamDial, a large-scale multi-domain dataset covering Automotive, Restaurant, and Hotel. StreamDial contains 87,498 dialogue sessions and 1,497,320 turns in total, with an average of 17.11 turns per session and a comparable scale across domains. Each session is organized as a structured quadruplet $\langle P_u, P_a, B, H \rangle$ that pairs dialogue history with explicit user/agent personas and a Conversational Blueprint, capturing realistic service behaviors such as requirement mining, constraint conflicts, negotiation, and recovery. Evaluations with automatic judges and downstream tasks show that StreamDial improves intrinsic dialogue quality over strong baselines, and models trained with StreamDial improve Dialogue State Tracking across backbones; we further report a completed human-evaluation set and encouraging multilingual transfer on Qwen3-8B under a controlled training budget. The data is released in https://github.com/hitxueliang/DialogDataSetBySTREAM.

cs.CL↗

arXiv2025

MME-RAG: Multi-Manager-Expert Retrieval-Augmented Generation for Fine-Grained Entity Recognition in Task-Oriented Dialogues

Fine-grained entity recognition is crucial for reasoning and decision-making in task-oriented dialogues, yet current large language models (LLMs) continue to face challenges in domain adaptation and retrieval controllability. We introduce MME-RAG, a Multi-Manager-Expert Retrieval-Augmented Generation framework that decomposes entity recognition into two coordinated stages: type-level judgment by lightweight managers and span-level extraction by specialized experts. Each expert is supported by a KeyInfo retriever that injects semantically aligned, few-shot exemplars during inference, enabling precise and domain-adaptive extraction without additional training. Experiments on CrossNER, MIT-Movie, MIT-Restaurant, and our newly constructed multi-domain customer-service dataset demonstrate that MME-RAG performs better than recent baselines in most domains. Ablation studies further show that both the hierarchical decomposition and KeyInfo-guided retrieval are key drivers of robustness and cross-domain generalization, establishing MME-RAG as a scalable and interpretable solution for adaptive dialogue understanding.

cs.CL↗

arXiv2025

Privacy-Preserving Explainable AIoT Application via SHAP Entropy Regularization

The widespread integration of Artificial Intelligence of Things (AIoT) in smart home environments has amplified the demand for transparent and interpretable machine learning models. To foster user trust and comply with emerging regulatory frameworks, the Explainable AI (XAI) methods, particularly post-hoc techniques such as SHapley Additive exPlanations (SHAP), and Local Interpretable Model-Agnostic Explanations (LIME), are widely employed to elucidate model behavior. However, recent studies have shown that these explanation methods can inadvertently expose sensitive user attributes and behavioral patterns, thereby introducing new privacy risks. To address these concerns, we propose a novel privacy-preserving approach based on SHAP entropy regularization to mitigate privacy leakage in explainable AIoT applications. Our method incorporates an entropy-based regularization objective that penalizes low-entropy SHAP attribution distributions during training, promoting a more uniform spread of feature contributions. To evaluate the effectiveness of our approach, we developed a suite of SHAP-based privacy attacks that strategically leverage model explanation outputs to infer sensitive information. We validate our method through comparative evaluations using these attacks alongside utility metrics on benchmark smart home energy consumption datasets. Experimental results demonstrate that SHAP entropy regularization substantially reduces privacy leakage compared to baseline models, while maintaining high predictive accuracy and faithful explanation fidelity. This work contributes to the development of privacy-preserving explainable AI techniques for secure and trustworthy AIoT applications.

cs.CR↗

arXiv2025

Enhancing Adversarial Robustness of IoT Intrusion Detection via SHAP-Based Attribution Fingerprinting

The rapid proliferation of Internet of Things (IoT) devices has transformed numerous industries by enabling seamless connectivity and data-driven automation. However, this expansion has also exposed IoT networks to increasingly sophisticated security threats, including adversarial attacks targeting artificial intelligence (AI) and machine learning (ML)-based intrusion detection systems (IDS) to deliberately evade detection, induce misclassification, and systematically undermine the reliability and integrity of security defenses. To address these challenges, we propose a novel adversarial detection model that enhances the robustness of IoT IDS against adversarial attacks through SHapley Additive exPlanations (SHAP)-based fingerprinting. Using SHAP's DeepExplainer, we extract attribution fingerprints from network traffic features, enabling the IDS to reliably distinguish between clean and adversarially perturbed inputs. By capturing subtle attribution patterns, the model becomes more resilient to evasion attempts and adversarial manipulations. We evaluated the model on a standard IoT benchmark dataset, where it significantly outperformed a state-of-the-art method in detecting adversarial attacks. In addition to enhanced robustness, this approach improves model transparency and interpretability, thereby increasing trust in the IDS through explainable AI.

cs.CR↗

arXiv2024

SEW: Self-calibration Enhanced Whole Slide Pathology Image Analysis

Pathology images are considered the ``gold standard" for cancer diagnosis and treatment, with gigapixel images providing extensive tissue and cellular information. Existing methods fail to simultaneously extract global structural and local detail features for comprehensive pathology image analysis efficiently. To address these limitations, we propose a self-calibration enhanced framework for whole slide pathology image analysis, comprising three components: a global branch, a focus predictor, and a detailed branch. The global branch initially classifies using the pathological thumbnail, while the focus predictor identifies relevant regions for classification based on the last layer features of the global branch. The detailed extraction branch then assesses whether the magnified regions correspond to the lesion area. Finally, a feature consistency constraint between the global and detail branches ensures that the global branch focuses on the appropriate region and extracts sufficient discriminative features for final identification. These focused discriminative features prove invaluable for uncovering novel prognostic tumor markers from the perspective of feature cluster uniqueness and tissue spatial distribution. Extensive experiment results demonstrate that the proposed framework can rapidly deliver accurate and explainable results for pathological grading and prognosis tasks.

cs.CV↗

arXiv2024

Benchmarking Multi-Scene Fire and Smoke Detection

The current irregularities in existing public Fire and Smoke Detection (FSD) datasets have become a bottleneck in the advancement of FSD technology. Upon in-depth analysis, we identify the core issue as the lack of standardized dataset construction, uniform evaluation systems, and clear performance benchmarks. To address this issue and drive innovation in FSD technology, we systematically gather diverse resources from public sources to create a more comprehensive and refined FSD benchmark. Additionally, recognizing the inadequate coverage of existing dataset scenes, we strategically expand scenes, relabel, and standardize existing public FSD datasets to ensure accuracy and consistency. We aim to establish a standardized, realistic, unified, and efficient FSD research platform that mirrors real-life scenes closely. Through our efforts, we aim to provide robust support for the breakthrough and development of FSD technology. The project is available at \href{https://xiaoyihan6.github.io/FSD/}{https://xiaoyihan6.github.io/FSD/}.

cs.CV↗

arXiv2024

When Swarm Learning meets energy series data: A decentralized collaborative learning design based on blockchain

Machine learning models offer the capability to forecast future energy production or consumption and infer essential unknown variables from existing data. However, legal and policy constraints within specific energy sectors render the data sensitive, presenting technical hurdles in utilizing data from diverse sources. Therefore, we propose adopting a Swarm Learning (SL) scheme, which replaces the centralized server with a blockchain-based distributed network to address the security and privacy issues inherent in Federated Learning (FL)'s centralized architecture. Within this distributed Collaborative Learning framework, each participating organization governs nodes for inter-organizational communication. Devices from various organizations utilize smart contracts for parameter uploading and retrieval. Consensus mechanism ensures distributed consistency throughout the learning process, guarantees the transparent trustworthiness and immutability of parameters on-chain. The efficacy of the proposed framework is substantiated across three real-world energy series modeling scenarios with superior performance compared to Local Learning approaches, simultaneously emphasizing enhanced data security and privacy over Centralized Learning and FL method. Notably, as the number of data volume and the count of local epochs increases within a threshold, there is an improvement in model performance accompanied by a reduction in the variance of performance errors. Consequently, this leads to an increased stability and reliability in the outcomes produced by the model.

cs.LG↗

arXiv2024

Joint Association, Beamforming, and Resource Allocation for Multi-IRS Enabled MU-MISO Systems With RSMA

Intelligent reflecting surface (IRS) and rate-splitting multiple access (RSMA) technologies are at the forefront of enhancing spectrum and energy efficiency in the next generation multi-antenna communication systems. This paper explores a RSMA system with multiple IRSs, and proposes two purpose-driven scheduling schemes, i.e., the exhaustive IRS-aided (EIA) and opportunistic IRS-aided (OIA) schemes. The aim is to optimize the system weighted energy efficiency (EE) under the above two schemes, respectively. Specifically, the Dinkelbach, branch and bound, successive convex approximation, and the semidefinite relaxation methods are exploited within the alternating optimization framework to obtain effective solutions to the considered problems. The numerical findings indicate that the EIA scheme exhibits better performance compared to the OIA scheme in diverse scenarios when considering the weighted EE, and the proposed algorithm demonstrates superior performance in comparison to the baseline algorithms.

eess.SP↗

arXiv2023

Sentence Bag Graph Formulation for Biomedical Distant Supervision Relation Extraction

We introduce a novel graph-based framework for alleviating key challenges in distantly-supervised relation extraction and demonstrate its effectiveness in the challenging and important domain of biomedical data. Specifically, we propose a graph view of sentence bags referring to an entity pair, which enables message-passing based aggregation of information related to the entity pair over the sentence bag. The proposed framework alleviates the common problem of noisy labeling in distantly supervised relation extraction and also effectively incorporates inter-dependencies between sentences within a bag. Extensive experiments on two large-scale biomedical relation datasets and the widely utilized NYT dataset demonstrate that our proposed framework significantly outperforms the state-of-the-art methods for biomedical distant supervision relation extraction while also providing excellent performance for relation extraction in the general text mining domain.

cs.LG↗

arXiv2023

Physics Driven Deep Retinex Fusion for Adaptive Infrared and Visible Image Fusion

Convolutional neural networks have turned into an illustrious tool for image fusion and super-resolution. However, their excellent performance cannot work without large fixed-paired datasets; and additionally, these high-demanded ground truth data always cannot be obtained easily in fusion tasks. In this study, we show that, the structures of generative networks capture a great deal of image feature priors, and then these priors are sufficient to reconstruct high-quality fused super-resolution result using only low-resolution inputs. By this way, we propose a novel self-supervised dataset-free method for adaptive infrared (IR) and visible (VIS) image super-resolution fusion named Deep Retinex Fusion (DRF). The key idea of DRF is first generating component priors which are disentangled from physical model using our designed generative networks ZipperNet, LightingNet and AdjustingNet, then combining these priors which captured by networks via adaptive fusion loss functions based on Retinex theory, and finally reconstructing the super-resolution fusion results. Furthermore, in order to verify the effectiveness of our reported DRF, both qualitative and quantitative experiments via comparing with other state-of-the-art methods are performed using different test sets. These results prove that, comparing with large datasets trained methods, DRF which works without any dataset achieves the best super-resolution fusion performance; and more importantly, DRF can adaptively balance IR and VIS information and has good noise immunity. DRF codes are open source available at https://github.com/GuYuanjie/Deep-Retinex-fusion.

cs.CV↗

arXiv2021

Document Layout Analysis with Aesthetic-Guided Image Augmentation

Document layout analysis (DLA) plays an important role in information extraction and document understanding. At present, document layout analysis has reached a milestone achievement, however, document layout analysis of non-Manhattan is still a challenge. In this paper, we propose an image layer modeling method to tackle this challenge. To measure the proposed image layer modeling method, we propose a manually-labeled non-Manhattan layout fine-grained segmentation dataset named FPD. As far as we know, FPD is the first manually-labeled non-Manhattan layout fine-grained segmentation dataset. To effectively extract fine-grained features of documents, we propose an edge embedding network named L-E^3Net. Experimental results prove that our proposed image layer modeling method can better deal with the fine-grained segmented document of the non-Manhattan layout.

cs.CV↗

arXiv2020

An efficient and high-resolution topology optimization method based on convolutional neural networks

2. In Section 3, we used some vague statements to affirm the training process of the neural network, which cannot support others to reproduce the results of the paper. In addition, this section does not show the difference between this paper and other work, nor does it reflect innovation. 3. In Section 5, the numerical examples are not compared with other multi-resolution methods, which is not enough to explain the superiority of the method proposed in this paper. Figure 10 also fails to show that this method is significantly improved compared to the traditional method. Thanks for your understanding. Topology optimization is a pioneering design method that can provide various candidates with high mechanical properties. However, the high-resolution for the optimum structures is highly desired, normally in turn leading to computationally intractable puzzle, especially for the famous Solid Isotropic Material with Penalization (SIMP) method. In this paper, we introduce the Super-Resolution Convolutional Neural Network (SRCNN) technique into topology optimization framework to improve the resolution of topology solutions with extremely high computational efficiency. Additionally, a pooling strategy is established to balance the number of finite element analysis (FEA) and the output mesh in optimization process. Considering the high training cost of 3D neural networks, several 2D neural networks are combined to deal with 3D topology optimization design problems. The combined treatment method used in 3D topology optimization design eliminates the expense of retraining 3D convolutional neural network and guarantees the quality of 3D design. Some typical examples justify that the high-resolution topology optimization method adopting SRCNN has excellent applicability and high efficiency.

math.OC↗

arXiv2019

Characteristics of Long Gamma-Ray bursts in the Comoving Frame

We compile a sample of 93 long gamma-ray bursts (GRBs) from Fermi satellite and 131 from Konus-Wind, which have measured redshifts and well determined spectra, and estimate their pseudo Lorentz factors (${Γ_0}$) using the tight ${L_{\rm{iso}}}$-${E_{\rm p}}$-${Γ_0}$ correlation. The statistical properties and pair correlations of temporal and spectral parameters are studied in the observer frame, rest frame and comoving frame, respectively. We find that the distributions of the duration, peak energy, isotropic energy and luminosity in the different frames are basically lognormal, and their distributions in the comoving frame are narrow, clustering around $T'_{\rm 90}\sim 4000$ s, $E'_{\rm p,c}\sim 0.7$ keV, $E'_{\rm iso,c} \sim 8\times10^{49}$ erg and $L'_{\rm iso,c}\sim 2.5\times10^{46}$ erg s$^{-1}$, where the redshift evolution effect has been taken into account. We also find that the values of ${Γ_0}$ are broadly distributed between few tens and several hundreds with median values $\sim 270$. We further analyze the pair correlations of all the quantities, and well confirm ${E_{\rm{iso}}}$-${E_{\rm p}}$, ${L_{\rm{iso}}}$-${E_{\rm p}}$, ${L_{\rm{iso}}}$-${Γ_0}$ and $E_{\rm{iso}}$-${Γ_0}$ relations, and find that the corresponding relations in the comoving frame do still exist, but have large dispersions. This suggests not only the well-known spectrum-energy relations are intrinsic correlations, but also the observed correlations are governed by the Doppler effect. In addition, the peak energies of long GRBs are independent of durations both in the rest frame and in the comoving frame. And there is a weak anticorrelation between the peak energy and Lorentz factor.

astro-ph.HE↗

arXiv2018

Assured Data Deletion with Fine-grained Access Control for Fog-based Industrial Applications

The advances of cloud computing, fog computing and Internet of Things (IoT) make the industries more prosperous than ever. A wide range of industrial systems such as transportation systems and manufacturing systems have been developed by integrating cloud computing, fog computing and IoT successfully. Security and privacy issues are a major concern that hinders the wide adoptions of these novel techniques. In this paper, we focus on assured data deletion, an issue which is important but received less attention in academia and industry. We firstly propose a framework to integrate the cloud, the fog and the things together to manage the stored data from industries or individuals. We then focus on secure data deletion in this framework by proposing an assured data deletion scheme which fulfills fine-grained access control over sensitive data and verifiable data deletion. Only the data owners and the fog devices are involved when deleting a data key and validating the data deletion, which makes the protocol practical due to the features of low latency and real-time interaction of fog computing. The proposed protocol takes advantage of attribute-based encryption and is provably secure under the standard model. The theoretical analysis shows the good performance and functionality requirements while the implementation results demonstrate the feasibility of our proposal.

cs.CR↗

arXiv2014

Cross-Layer Scheduling for OFDMA-based Cognitive Radio Systems with Delay and Security Constraints

This paper considers the resource allocation problem in an Orthogonal Frequency Division Multiple Access (OFDMA) based cognitive radio (CR) network, where the CR base station adopts full overlay scheme to transmit both private and open information to multiple users with average delay and power constraints. A stochastic optimization problem is formulated to develop flow control and radio resource allocation in order to maximize the long-term system throughput of open and private information in CR system and ensure the stability of primary system. The corresponding optimal condition for employing full overlay is derived in the context of concurrent transmission of open and private information. An online resource allocation scheme is designed to adapt the transmission of open and private information based on monitoring the status of primary system as well as the channel and queue states in the CR network. The scheme is proven to be asymptotically optimal in solving the stochastic optimization problem without knowing any statistical information. Simulations are provided to verify the analytical results and efficiency of the scheme.

cs.NI↗

arXiv2012

Anisotropy as a Probe of the Galactic Cosmic Ray Propagation and the Galactic Halo Magnetic Field

The anisotropy of cosmic rays (CRs) in the solar vicinity is generally at- tributed to the CR streaming due to the discrete distribution of CR sources or local magnetic field modulation. Recently, the two dimensional large scale CR anisotropy has been measured by many experiments in TeV-PeV energy range in both hemispheres. The tail-in excess along the tangential direction of the local spiral arm and the loss cone deficit pointing to the north Galactic pole direction agree with what have been obtained in tens to hundreds of GeV. The persistence of the two large scale anisotropy structures in such a wide range of energy suggests that the anisotropy might be due to a global streaming of the Galactic CRs (GCRs). This work tries to extend the observed CR anisotropy picture from solar system to the whole galaxy. In such a case, we can find a new interesting signature, a loop of GCR streaming, of the GCR propagation. We further calculate the overall GCR streaming induced magnetic field, and find a qualitative consistence with the observed structure of the halo magnetic field.

astro-ph.GA↗

arXiv2011

Observation of the antimatter helium-4 nucleus at RHIC

We present the observation of the \Heebar nucleus, the heaviest antinucleus observed to date. In total, 18 \Heebar counts were detected at the STAR experiment at RHIC in 10$^{9}$ recorded Au+Au collisions at beam energies of $\sqrt{s_{NN}}$ = 200 GeV and 62 GeV. The background has been estimated, and the misidentification probability is found to be lower than 10$^{-11}$.

nucl-ex↗