Xuanrong Huo

Xuanrong Huo

Auto-generated profile, based on publicly available information

Xuanrong Huo

School of Computer, Qinghai Normal University, Xining, PR China

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 April 2026
Verified
Conference Paper
Open access
HWHaoxiang Wen
Haoxiang Wen
XGXiaolei Gu
Xiaolei Gu
JHJiaxin Han
Jiaxin Han
XHXuanrong Huo
Xuanrong Huo
ZYZhonglin Ye
Zhonglin Ye
WLWen Lu
Wen Lu

27 July 2026 in Nature Communications
Verified
Journal Article
Open access
PDF file
JHJiaxin Han
Jiaxin Han
XHXuanrong Huo
Xuanrong Huo
YXYuzhi Xiao
Yuzhi Xiao
ZYZhonglin Ye
Zhonglin Ye
Yuhui ZhengYuhui Zheng
Yuhui Zheng
ZLZhen Liu
Zhen Liu
Existing multivariate anomaly detection methods suffer from practical industrial limitations stemming from distribution assumptions, volatile data and scarce reliable anomaly labels. This study proposes a hyperbolic adaptive spatial-aware multivariate anomaly detection method aimed at enhancing accuracy, robustness and interpretability in real-world applications. It first constructs an adaptive hyperbolic pre-training framework to embed coupled spatio-temporal hierarchical time-series features and mine latent cross-variable correlations. Next, an adaptive graph structure discovery and dynamic sparsification module removes rigid topology restraints, autonomously learning inherent data structures and pinpointing the spatio-temporal locations of anomalies for causal interpretation. A multi-level attention module and small-sample dynamic threshold algorithm further improve model stability when handling complex signals. Built upon self-adaptive graph topology, the unified end-to-end framework integrates anomaly recognition, diagnosis and interpretation with dedicated discrimination and correction mechanisms. Experiments demonstrate 10%–30% performance improvements across diverse detection benchmarks, verifying the method’s effectiveness and advancing interpretable AI research for anomaly detection.

3 April 2026
Verified
Conference Paper
Open access
ZWZixu Wang
Zixu Wang
XGXiaolei Gu
Xiaolei Gu
WLWen Lu
Wen Lu
ZYZhonglin Ye
Zhonglin Ye
XHXuanrong Huo
Xuanrong Huo
In recent years, the rapid advancement of computing power networks and computing-network integration has made the coordinated scheduling of network and computing resources a critical component of next-generation information infrastructure. Resource scheduling in cloud-edge-end collaborative environments exhibits high heterogeneity, strong coupling, and dynamic complexity. This challenges traditional optimization methods, such as mathematical programming and heuristic algorithms, by introducing modeling difficulties and real-time limitations in large-scale, complex scenarios. However, the introduction of reinforcement learning, graph neural networks, and large models into computing network resource scheduling has successfully advanced scheduling mechanisms toward intelligence, adaptability, and cross-scenario generalization. This paper systematically reviews optimal allocation methods for computing network resources, categorizes and summarizes research progress in traditional optimization methods, reinforcement learning approaches, graph neural network methods, and large-model techniques, and compares the capability boundaries of these methods. Finally, it discusses current research challenges and future development trends.