Canqun Yang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Canqun Yang has a total of 120 co-authors. Here are the top co-authors by number of shared publications:
- Xiao-Wei Guo
National University of Defense Technology
8 shared publications - Tao Tang
National University of Defense Technology
6 shared publications - Chao Li
School of Computer Science and Technology, National University of Defense Technology, Changsha, China
5 shared publications - Jianbin Fang
College of Computer Science and Technology, National University of Defense Technology, Changsha, China
5 shared publications - Sen Zhang
National University of Defense Technology, China
5 shared publications - Chun Huang
College of Computer Science and Technology, National University of Defense Technology, Changsha, China
4 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 15216-15225
Preserving Geometric Fidelity in Efficient Transformer-Based PDE Solvers
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 15216-15225
Open AccessPreserving Geometric Fidelity in Efficient Transformer-Based PDE Solvers
- 1 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (3) , 3652-3664
Temporal-Enhanced Multimodal Transformer for Referring Multi-Object Tracking and Segmentation
- 1 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (3) , 3652-3664
Temporal-Enhanced Multimodal Transformer for Referring Multi-Object Tracking and Segmentation
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (9) , 1-14
Optimizing Long-read Sequence Alignment on a CPU-DSPs Heterogeneous Processor
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (9) , 1-14
Optimizing Long-read Sequence Alignment on a CPU-DSPs Heterogeneous Processor
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 2007-2015
MMF-SV: A Multi-Modal Feature Fusion-Based Structural Variant Caller
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 2007-2015
Open AccessMMF-SV: A Multi-Modal Feature Fusion-Based Structural Variant Caller
- 22 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 33 (3) , 1326-1337
Real-Time Anomaly Detection for Large-Scale Network Devices
With the booming of large-scale network devices, anomaly detection on multivariate time series (MTS), such as a combination of CPU utilization, average response time, and network packet loss, is important for system reliability. Although a collection of learning-based approaches have been designed for this purpose, our study shows that these approaches suffer from long initialization time for sufficient training data. Our previously proposed JumpStarter model stands as a MTS anomaly detection method characterized by its brief initialization time and commendable detection performance. However, it suffers from high computational cost and inappropriateness for periodic MTS. In this paper, we propose VersaGuardian, which introduces the Dynamic Mode Decomposition technique to MTS anomaly detection for diverse types of MTS in a rapidly initialized, computationally efficient manner. With real-world MTS datasets collected from three companies, our results show that VersaGuardian achieves an average F1 score of 94.42%, significantly outperforming the popular anomaly detection algorithms, with a much shorter initialization time of 20 minutes and detection time of 15.28 milliseconds.
- 22 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 33 (3) , 1326-1337
Real-Time Anomaly Detection for Large-Scale Network Devices
With the booming of large-scale network devices, anomaly detection on multivariate time series (MTS), such as a combination of CPU utilization, average response time, and network packet loss, is important for system reliability. Although a collection of learning-based approaches have been designed for this purpose, our study shows that these approaches suffer from long initialization time for sufficient training data. Our previously proposed JumpStarter model stands as a MTS anomaly detection method characterized by its brief initialization time and commendable detection performance. However, it suffers from high computational cost and inappropriateness for periodic MTS. In this paper, we propose VersaGuardian, which introduces the Dynamic Mode Decomposition technique to MTS anomaly detection for diverse types of MTS in a rapidly initialized, computationally efficient manner. With real-world MTS datasets collected from three companies, our results show that VersaGuardian achieves an average F1 score of 94.42%, significantly outperforming the popular anomaly detection algorithms, with a much shorter initialization time of 20 minutes and detection time of 15.28 milliseconds.
- Open Access
- 28 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 24-30
Discrete-time Physics-Informed Neural Networks for Two-Phase Flow Interface Capturing
In two-phase flow, accurately capturing sharp interfaces is essential. Physics-Informed Neural Networks (PINNs) provide a new way to capture interfaces. This paper introduces a new framework that uses Discrete-time Physics-Informed Neural Networks (DtPINNs) to solve the volume of fluid (VOF) advection equation, offering a new approach to interface capturing in two-phase flows. With this framework, we propose an adaptive collocation-point refinement algorithm, which improves the precision of capturing sharp interfaces, reducing errors and diffusion. Experiments, including cases with translation and deformation, show that the DtPINNs method maintains sharp interfaces, outperforming traditional numerical methods. In a 2D single-vortex deformation case, DtPINNs achieved a relative \({L}_2\) error of 0.0760, much lower than the 5th-order WENO-JS scheme (0.5534) and the 1st-order Upwind scheme (0.8879). This shows that DtPINNs are better at capturing complex interface shapes while keeping accuracy and minimizing numerical diffusion.- 28 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 24-30
Open AccessDiscrete-time Physics-Informed Neural Networks for Two-Phase Flow Interface Capturing
In two-phase flow, accurately capturing sharp interfaces is essential. Physics-Informed Neural Networks (PINNs) provide a new way to capture interfaces. This paper introduces a new framework that uses Discrete-time Physics-Informed Neural Networks (DtPINNs) to solve the volume of fluid (VOF) advection equation, offering a new approach to interface capturing in two-phase flows. With this framework, we propose an adaptive collocation-point refinement algorithm, which improves the precision of capturing sharp interfaces, reducing errors and diffusion. Experiments, including cases with translation and deformation, show that the DtPINNs method maintains sharp interfaces, outperforming traditional numerical methods. In a 2D single-vortex deformation case, DtPINNs achieved a relative \({L}_2\) error of 0.0760, much lower than the 5th-order WENO-JS scheme (0.5534) and the 1st-order Upwind scheme (0.8879). This shows that DtPINNs are better at capturing complex interface shapes while keeping accuracy and minimizing numerical diffusion. - Open Access
- 27 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1107-1119
Giving Every Modality a Voice in Microservice Failure Diagnosis via Multimodal Adaptive Optimization
Microservice systems are inherently complex and prone to failures, which can significantly impact user experience. Existing diagnostic approaches based on single-modal data such as logs, metrics, or traces cannot comprehensively capture failure patterns. For those multimodal data-based failure diagnosis methods, the dominant modality can overshadow others, hindering low-yield modalities from fully leveraging their characteristics. This paper proposes Medicine, a modal-independent microservice failure diagnosis framework based on multimodal adaptive optimization. It encodes different modalities separately to retain their unique features and employs adaptive optimization to adjust the learning pace between modalities, thereby enhancing overall diagnostic performance. Experimental results demonstrate that Medicine outperforms existing single-modal and multimodal diagnostic approaches on three public datasets, with F1-score improving by 15.72% to 70.84%. Even in cases where individual modal data is missing or of lower quality, Medicine maintains high diagnostic accuracy.
- 27 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1107-1119
Open AccessGiving Every Modality a Voice in Microservice Failure Diagnosis via Multimodal Adaptive Optimization
Microservice systems are inherently complex and prone to failures, which can significantly impact user experience. Existing diagnostic approaches based on single-modal data such as logs, metrics, or traces cannot comprehensively capture failure patterns. For those multimodal data-based failure diagnosis methods, the dominant modality can overshadow others, hindering low-yield modalities from fully leveraging their characteristics. This paper proposes Medicine, a modal-independent microservice failure diagnosis framework based on multimodal adaptive optimization. It encodes different modalities separately to retain their unique features and employs adaptive optimization to adjust the learning pace between modalities, thereby enhancing overall diagnostic performance. Experimental results demonstrate that Medicine outperforms existing single-modal and multimodal diagnostic approaches on three public datasets, with F1-score improving by 15.72% to 70.84%. Even in cases where individual modal data is missing or of lower quality, Medicine maintains high diagnostic accuracy.
- Open Access
Temporal-Enhanced Multimodal Transformer for Referring Multi-Object Tracking and Segmentation
Referring multi-object tracking (RMOT) is an emerging cross-modal task that aims to locate an arbitrary number of target objects and maintain their identities referred by a language expression in a video. This intricate task involves the reasoning of linguistic and visual modalities, along with the temporal association of target objects. However, the seminal work employs only loose feature fusion and overlooks the utilization of long-term information on tracked objects. In this study, we introduce a compact Transformer-based method, termed TenRMOT. We conduct feature fusion at both encoding and decoding stages to fully exploit the advantages of Transformer architecture. Specifically, we incrementally perform cross-modal fusion layer-by-layer during the encoding phase. In the decoding phase, we utilize language-guided queries to probe memory features for accurate prediction of the desired objects. Moreover, we introduce a query update module that explicitly leverages temporal prior information of the tracked objects to enhance the consistency of their trajectories. In addition, we introduce a novel task called Referring Multi-Object Tracking and Segmentation (RMOTS) and construct a new dataset named Ref-KITTI Segmentation. Our dataset consists of 18 videos with 818 expressions, and each expression averages 10.7 masks, which poses a greater challenge compared to the typical single mask in most existing referring video segmentation datasets. TenRMOT demonstrates superior performance on both the referring multi-object tracking and the segmentation tasks.
Open AccessTemporal-Enhanced Multimodal Transformer for Referring Multi-Object Tracking and Segmentation
Referring multi-object tracking (RMOT) is an emerging cross-modal task that aims to locate an arbitrary number of target objects and maintain their identities referred by a language expression in a video. This intricate task involves the reasoning of linguistic and visual modalities, along with the temporal association of target objects. However, the seminal work employs only loose feature fusion and overlooks the utilization of long-term information on tracked objects. In this study, we introduce a compact Transformer-based method, termed TenRMOT. We conduct feature fusion at both encoding and decoding stages to fully exploit the advantages of Transformer architecture. Specifically, we incrementally perform cross-modal fusion layer-by-layer during the encoding phase. In the decoding phase, we utilize language-guided queries to probe memory features for accurate prediction of the desired objects. Moreover, we introduce a query update module that explicitly leverages temporal prior information of the tracked objects to enhance the consistency of their trajectories. In addition, we introduce a novel task called Referring Multi-Object Tracking and Segmentation (RMOTS) and construct a new dataset named Ref-KITTI Segmentation. Our dataset consists of 18 videos with 818 expressions, and each expression averages 10.7 masks, which poses a greater challenge compared to the typical single mask in most existing referring video segmentation datasets. TenRMOT demonstrates superior performance on both the referring multi-object tracking and the segmentation tasks.
- Open Access
Event-Oriented State Alignment Network for Weakly Supervised Temporal Language Grounding
Weakly supervised temporal language grounding (TLG) aims to locate events in untrimmed videos based on natural language queries without temporal annotations, necessitating a deep understanding of semantic context across both video and text modalities. Existing methods often focus on simple correlations between query phrases and isolated video segments, neglecting the event-oriented semantic coherence and consistency required for accurate temporal grounding. This can lead to misleading results due to partial frame correlations. To address these limitations, we propose the Event-oriented State Alignment Network (ESAN), which constructs “start–event–end” semantic state sets for both textual and video data. ESAN employs relative entropy for cross-modal alignment through knowledge distillation from pre-trained large models, thereby enhancing semantic coherence within each modality and ensuring consistency across modalities. Our approach leverages vision–language models to extract static frame semantics and large language models to capture dynamic semantic changes, facilitating a more comprehensive understanding of events. Experiments conducted on two benchmark datasets demonstrate that ESAN significantly outperforms existing methods. By reducing false high correlations and improving the overall performance, our method effectively addresses the challenges posed by previous approaches. These advancements highlight the potential of ESAN to improve the precision and reliability of temporal language grounding tasks.
Open AccessEvent-Oriented State Alignment Network for Weakly Supervised Temporal Language Grounding
Weakly supervised temporal language grounding (TLG) aims to locate events in untrimmed videos based on natural language queries without temporal annotations, necessitating a deep understanding of semantic context across both video and text modalities. Existing methods often focus on simple correlations between query phrases and isolated video segments, neglecting the event-oriented semantic coherence and consistency required for accurate temporal grounding. This can lead to misleading results due to partial frame correlations. To address these limitations, we propose the Event-oriented State Alignment Network (ESAN), which constructs “start–event–end” semantic state sets for both textual and video data. ESAN employs relative entropy for cross-modal alignment through knowledge distillation from pre-trained large models, thereby enhancing semantic coherence within each modality and ensuring consistency across modalities. Our approach leverages vision–language models to extract static frame semantics and large language models to capture dynamic semantic changes, facilitating a more comprehensive understanding of events. Experiments conducted on two benchmark datasets demonstrate that ESAN significantly outperforms existing methods. By reducing false high correlations and improving the overall performance, our method effectively addresses the challenges posed by previous approaches. These advancements highlight the potential of ESAN to improve the precision and reliability of temporal language grounding tasks.
- 26 August 2024
- Published by Springer Nature
- p. 371-385
VLASPH: Smoothed Particle Hydrodynamics on VLA SIMD Architectures
- 26 August 2024
- Published by Springer Nature
- p. 371-385
VLASPH: Smoothed Particle Hydrodynamics on VLA SIMD Architectures