Yuchen 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
Yuchen Yang has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Hongli Zhang
School of Cyberspace Science, Harbin Institute of Technology, Harbin, China
9 shared publications - Gongzhu Yin
School of Cyberspace Science, Harbin Institute of Technology, Harbin, China
8 shared publications - Chao Meng
Faculty of Computing, Harbin Institute of Technology, Harbin, China
4 shared publications - Kun Lu
School of Cyberspace Science, Harbin Institute of Technology, Harbin, China
4 shared publications - Yi Luo
School of Cyberspace Science, Faculty of Computing, Harbin Institute of Technology, Harbin 150001, China
4 shared publications Binxing Fang
2 shared publications
Latest Publications
- Open Access
- 1 July 2026
- Published by Elsevier in SSRN Electronic Journal
Causally Motivated Mechanism Aggregation for Shift-Robust Tabular Anomaly Detection
Tabular anomaly detection is central to structured decision systems in finance, healthcare, and industrial monitoring, where undetected anomalies can indicate fraud, disease, or system failure.Most deep detectors learn correlations in the training data, which can make their anomaly scores unstable under distribution shift. They can also be weak for hidden mechanism violations that remain statistically plausible: local feature relationships look normal, but the coordinated cross-variable changes required by the underlying mechanism are inconsistent.Causal representation learning offers a principled route to robustness, but existing methods usually require ground-truth graphs, interventions, or strong structural priors that are rarely available for high-dimensional tabular data. Our key insight is that anomaly detection can exploit mechanism-level structure without recovering a complete causal graph, because anomalies often disrupt groups of variables governed by the same generative mechanism.We propose PRISM-VAE, a variational framework that learns latent mechanism blocks by shaping local sensitivity in the representation space, without predefined causal graphs. This shifts from variable-level structure learning to mechanism aggregation, reducing overhead to $O(m)$ and making mechanism-level modeling practical for high-dimensional tabular applications. Across twelve real-world datasets, PRISM-VAE improves AUROC by up to 11.3 percentage points over strong baselines and shows stronger robustness under evaluated distribution shifts. Using a controlled synthetic financial fraud benchmark whose data-generating process specifies the ground-truth causal mechanisms, PRISM-VAE achieves mechanism aggregation without structural priors and remains competitive with methods given causal supervision. We also introduce a mechanism-level metric to test alignment between learned latent blocks and ground-truth mechanisms.- 1 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessCausally Motivated Mechanism Aggregation for Shift-Robust Tabular Anomaly Detection
Tabular anomaly detection is central to structured decision systems in finance, healthcare, and industrial monitoring, where undetected anomalies can indicate fraud, disease, or system failure.Most deep detectors learn correlations in the training data, which can make their anomaly scores unstable under distribution shift. They can also be weak for hidden mechanism violations that remain statistically plausible: local feature relationships look normal, but the coordinated cross-variable changes required by the underlying mechanism are inconsistent.Causal representation learning offers a principled route to robustness, but existing methods usually require ground-truth graphs, interventions, or strong structural priors that are rarely available for high-dimensional tabular data. Our key insight is that anomaly detection can exploit mechanism-level structure without recovering a complete causal graph, because anomalies often disrupt groups of variables governed by the same generative mechanism.We propose PRISM-VAE, a variational framework that learns latent mechanism blocks by shaping local sensitivity in the representation space, without predefined causal graphs. This shifts from variable-level structure learning to mechanism aggregation, reducing overhead to $O(m)$ and making mechanism-level modeling practical for high-dimensional tabular applications. Across twelve real-world datasets, PRISM-VAE improves AUROC by up to 11.3 percentage points over strong baselines and shows stronger robustness under evaluated distribution shifts. Using a controlled synthetic financial fraud benchmark whose data-generating process specifies the ground-truth causal mechanisms, PRISM-VAE achieves mechanism aggregation without structural priors and remains competitive with methods given causal supervision. We also introduce a mechanism-level metric to test alignment between learned latent blocks and ground-truth mechanisms. - Open Access
- 23 May 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (11) , 1804
RSCF-PM: Relation-Specific Curvature Fields on Product Manifolds for Fraud Detection in Multi-Relational Social Networks
Graph-based fraud detection in multi-relational social networks must capture heterogeneous relation semantics and diverse fraud patterns while preserving geometric consistency and remaining scalable. Existing methods often either force all relations into a shared Euclidean or single-curvature space, or fuse relation-wise embeddings after mapping them to tangent coordinates, which weakens curvature-dependent metric information. We propose Relation-Specific Curvature Fields on Product Manifolds (RSCF-PM), a geometry-consistent framework that learns relation-specific curvature and represents each node as a tuple on a Riemannian product manifold. Each relation is encoded in its own hyperbolic space, and cross-relation fusion is performed directly through the product metric rather than Euclidean concatenation. On top of this representation, we introduce a multi-prototype classifier to model multiple fraud modes within each class. To support large-scale training, we adopt tangent-space aggregation as an efficient approximation to the Fréchet mean. Experiments on four public fraud detection benchmarks, including the 5.78M-node T-Social network, show that RSCF-PM achieves the best results on T-Social, FDCompCN, and YelpChi, while remaining highly competitive on Amazon, with up to 4.96% AUC improvement over strong baselines. Ablation and efficiency studies further confirm the complementary value of each component and the practical scalability of the framework.
- 23 May 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (11) , 1804
Open AccessRSCF-PM: Relation-Specific Curvature Fields on Product Manifolds for Fraud Detection in Multi-Relational Social Networks
Graph-based fraud detection in multi-relational social networks must capture heterogeneous relation semantics and diverse fraud patterns while preserving geometric consistency and remaining scalable. Existing methods often either force all relations into a shared Euclidean or single-curvature space, or fuse relation-wise embeddings after mapping them to tangent coordinates, which weakens curvature-dependent metric information. We propose Relation-Specific Curvature Fields on Product Manifolds (RSCF-PM), a geometry-consistent framework that learns relation-specific curvature and represents each node as a tuple on a Riemannian product manifold. Each relation is encoded in its own hyperbolic space, and cross-relation fusion is performed directly through the product metric rather than Euclidean concatenation. On top of this representation, we introduce a multi-prototype classifier to model multiple fraud modes within each class. To support large-scale training, we adopt tangent-space aggregation as an efficient approximation to the Fréchet mean. Experiments on four public fraud detection benchmarks, including the 5.78M-node T-Social network, show that RSCF-PM achieves the best results on T-Social, FDCompCN, and YelpChi, while remaining highly competitive on Amazon, with up to 4.96% AUC improvement over strong baselines. Ablation and efficiency studies further confirm the complementary value of each component and the practical scalability of the framework.
- 2 January 2026
- Published by Springer Nature
- p. 746-756
Ignite Forecasting with SPARK: An Efficient Generative Framework for Refining LLMs in Temporal Knowledge Graph Forecasting
- 2 January 2026
- Published by Springer Nature
- p. 746-756
Ignite Forecasting with SPARK: An Efficient Generative Framework for Refining LLMs in Temporal Knowledge Graph Forecasting
- Open Access
- 1 January 2026
- Published by Springer Nature in World Wide Web
- Vol. 29 (3) , 27
MHGCA: Multi-scale heterogeneous graph learning with content-structure alignment for social spammer detection
- 1 January 2026
- Published by Springer Nature in World Wide Web
- Vol. 29 (3) , 27
Open AccessMHGCA: Multi-scale heterogeneous graph learning with content-structure alignment for social spammer detection
- Open Access
- 20 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1821-1832
Inductive Link Prediction on N-ary Relational Facts via Semantic Hypergraph Reasoning
- 20 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1821-1832
Open AccessInductive Link Prediction on N-ary Relational Facts via Semantic Hypergraph Reasoning
- Open Access
Ignite Forecasting with SPARK: An Efficient Generative Framework for Refining LLMs in Temporal Knowledge Graph Forecasting
Temporal Knowledge Graph (TKG) forecasting is crucial for predicting future events using historical data. With the surge of Large Language Models (LLMs), recent studies have begun exploring their integration into TKG forecasting and achieved some success. However, they still face limitations such as limited input length, inefficient output generation, and resource-intensive refinement, which undermine their performance and practical applicability. To address these limitations, we introduce SPARK, a Sequence-level Proxy-Adapting framework for Refining LLMs in TKG forecasting. Inspired by inference-time algorithms adopted in controlling generation, SPARK offers a cost-effective, plug-and-play solution through two key innovations: (1) Beam Sequence-Level Generation, which reframes TKG forecasting as a top-K sequence-level generation task, using beam search for efficiently generating next-entity distribution in a single forward pass. (2) TKG Adapter for Refinement, which employs traditional TKG models as trainable proxy adapters to leverage global graph information and refine LLM outputs, overcoming both the input length and the resource-intensive fine-tuning problems. Experiments across diverse datasets validate SPARK's forecasting performance, robust generalization capabilities, and high efficiency. We release source codes at https://github.com/yin-gz/SPARK.
Open AccessIgnite Forecasting with SPARK: An Efficient Generative Framework for Refining LLMs in Temporal Knowledge Graph Forecasting
Temporal Knowledge Graph (TKG) forecasting is crucial for predicting future events using historical data. With the surge of Large Language Models (LLMs), recent studies have begun exploring their integration into TKG forecasting and achieved some success. However, they still face limitations such as limited input length, inefficient output generation, and resource-intensive refinement, which undermine their performance and practical applicability. To address these limitations, we introduce SPARK, a Sequence-level Proxy-Adapting framework for Refining LLMs in TKG forecasting. Inspired by inference-time algorithms adopted in controlling generation, SPARK offers a cost-effective, plug-and-play solution through two key innovations: (1) Beam Sequence-Level Generation, which reframes TKG forecasting as a top-K sequence-level generation task, using beam search for efficiently generating next-entity distribution in a single forward pass. (2) TKG Adapter for Refinement, which employs traditional TKG models as trainable proxy adapters to leverage global graph information and refine LLM outputs, overcoming both the input length and the resource-intensive fine-tuning problems. Experiments across diverse datasets validate SPARK's forecasting performance, robust generalization capabilities, and high efficiency. We release source codes at https://github.com/yin-gz/SPARK.
- 22 March 2025
- Published by Springer Nature in Neural Computing and Applications
- Vol. 37 (18) , 11965-11990
Stance classification model with knowledge-aware multi-feature attention network
- 22 March 2025
- Published by Springer Nature in Neural Computing and Applications
- Vol. 37 (18) , 11965-11990
Stance classification model with knowledge-aware multi-feature attention network
- Open Access
- 7 October 2023
- Published by Atlantis Press SARL
- p. 1351-1365
Empirical Study of the Impact of Cross-Border Data Flow Restrictions on Digital Service Exports
- 7 October 2023
- Published by Atlantis Press SARL
- p. 1351-1365
Open AccessEmpirical Study of the Impact of Cross-Border Data Flow Restrictions on Digital Service Exports
- Open Access
- 27 February 2023
- Published by Association for Computing Machinery (ACM)
- p. 751-759
Beyond Individuals: Modeling Mutual and Multiple Interactions for Inductive Link Prediction between Groups
Link prediction is a core task in graph machine learning with wide applications. However, little attention has been paid to link prediction between two group entities. This limits the application of the current approaches to many real-life problems, such as predicting collaborations between academic groups or recommending bundles of items to group users. Moreover, groups are often ephemeral or emergent, forcing the predicting model to deal with challenging inductive scenes. To fill this gap, we develop a framework composed of a GNN-based encoder and neural-based aggregating networks, namely the Mutual Multi-view Attention Networks (MMAN). First, we adopt GNN-based encoders to model multiple interactions among members and groups through propagating. Then, we develop MMAN to aggregate members' node representations into multi-view group representations and compute the final results by pooling pairwise scores between views. Specifically, several view-guided attention modules are adopted when learning multi-view group representations, thus capturing diversified member weights and multifaceted group characteristics. In this way, MMAN can further mimic the mutual and multiple interactions between groups. We conduct experiments on three datasets, including two academic group link prediction datasets and one bundle-to-group recommendation dataset. The results demonstrate that the proposed approach can achieve superior performance on both tasks compared with plain GNN-based methods and other aggregating methods.
- 27 February 2023
- Published by Association for Computing Machinery (ACM)
- p. 751-759
Open AccessBeyond Individuals: Modeling Mutual and Multiple Interactions for Inductive Link Prediction between Groups
Link prediction is a core task in graph machine learning with wide applications. However, little attention has been paid to link prediction between two group entities. This limits the application of the current approaches to many real-life problems, such as predicting collaborations between academic groups or recommending bundles of items to group users. Moreover, groups are often ephemeral or emergent, forcing the predicting model to deal with challenging inductive scenes. To fill this gap, we develop a framework composed of a GNN-based encoder and neural-based aggregating networks, namely the Mutual Multi-view Attention Networks (MMAN). First, we adopt GNN-based encoders to model multiple interactions among members and groups through propagating. Then, we develop MMAN to aggregate members' node representations into multi-view group representations and compute the final results by pooling pairwise scores between views. Specifically, several view-guided attention modules are adopted when learning multi-view group representations, thus capturing diversified member weights and multifaceted group characteristics. In this way, MMAN can further mimic the mutual and multiple interactions between groups. We conduct experiments on three datasets, including two academic group link prediction datasets and one bundle-to-group recommendation dataset. The results demonstrate that the proposed approach can achieve superior performance on both tasks compared with plain GNN-based methods and other aggregating methods.
- Open Access
- 1 January 2020
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 62 (3) , 1053-1075
A Review of Data Cleaning Methods for Web Information System
Web information system (WIS) is frequently-used and indispensable in daily social life. WIS provides information services in many scenarios, such as electronic commerce, communities, and edutainment. Data cleaning plays an essential role in various WIS scenarios to improve the quality of data service. In this paper, we present a review of the state-of-the-art methods for data cleaning in WIS. According to the characteristics of data cleaning, we extract the critical elements of WIS, such as interactive objects, application scenarios, and core technology, to classify the existing works. Then, after elaborating and analyzing each category, we summarize the descriptions and challenges of data cleaning methods with sub-elements such as data & user interaction, data quality rule, model, crowdsourcing, and privacy preservation. Finally, we analyze various types of problems and provide suggestions for future research on data cleaning in WIS from the technology and interactive perspective.
- 1 January 2020
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 62 (3) , 1053-1075
Open AccessA Review of Data Cleaning Methods for Web Information System
Web information system (WIS) is frequently-used and indispensable in daily social life. WIS provides information services in many scenarios, such as electronic commerce, communities, and edutainment. Data cleaning plays an essential role in various WIS scenarios to improve the quality of data service. In this paper, we present a review of the state-of-the-art methods for data cleaning in WIS. According to the characteristics of data cleaning, we extract the critical elements of WIS, such as interactive objects, application scenarios, and core technology, to classify the existing works. Then, after elaborating and analyzing each category, we summarize the descriptions and challenges of data cleaning methods with sub-elements such as data & user interaction, data quality rule, model, crowdsourcing, and privacy preservation. Finally, we analyze various types of problems and provide suggestions for future research on data cleaning in WIS from the technology and interactive perspective.