Yue Liu

Yue Liu (刘悦) is a master student at degree College of Computer Science and Technology, National University of Defence Technology (NUDT). He is supervised by Prof. Xinwang Liu in Pattern Recognition & Machine Intelligence Lab (PRMI). His research interests include graph neural networks, deep clustering and self-supervised learning.

Email  /  Google Scholar  /  Twitter  /  Github  /  GF

• [2024.01] One paper has been accepted by IEEE T-KDE.
• [2024.01] Two papers have been accepted by ICLR 2024.
• [2024.01] One papers has been accepted by IEEE T-NNLS.
• [2023.12] Three papers have been accepted by AAAI 2024.
• [2023.12] One paper has been accepted by ICDE 2024.
• [2023.11] I won China National Scholarship for graduate students.
• [2023.09] One paper has been accepted by NeurIPS 2023.
• [2023.07] Four papers have been accepted by ACM MM 2023.
• [2023.07] One paper has been accepted by IEEE T-NNLS.
• [2023.06] One paper has been accepted by IEEE T-KDE.
• [2023.04] One paper has been accepted by ICML 2023.
• [2023.04] One paper has been accepted by IEEE T-NNLS.
• [2023.04] One paper has been accepted by SIGIR 2023.
• [2023.01] One paper has been accepted by ICLR 2023.
• [2022.12] I won China National Scholarship for graduate students.
• [2022.11] Three papers have been accepted by AAAI 2023.
• [2022.06] One paper has been accepted by ACM MM 2022.
• [2022.04] One paper has been accepted by IJCAI 2022.
• [2021.12] I won the first-class scholarship of freshman at NUDT.
• [2021.12] One paper has been accepted by AAAI 2022.
• [2020.12] I won China National Scholarship for undergraduate students.

More

My research interests mainly include self-supervised learning and graph neural networks. I'm also interested in their applications such as deep graph clustering, knowledge graph embedding, recommendation, bioinformatics, and LLMs.
* denotes equal contributions. The representative papers are highlighted.

We propose an intent learning method termed ELCRec, which leverages end-to-end learnable clustering and cluster-assisted contrastive learning to improve recommendation. Both the results on open benchmarks and industrial engine demonstrate the superiority.

We aim to extend deep graph clustering to temporal graphs, which are more practical in real-world scenarios. We propose a general framework TGC by clustering distribution assignment and adjacency reconstruction.

We explore at which training stage can code data help LLMs reasoning. The extensive experiments and insights deepen the understanding of LLMs' reasoning capability and the corresponding applications, e.g., scientific question answering, legal support, etc.

We propose RGC by determining the cluster number in deep graph clustering methods via the reinforcement learning.

We propose a plug-and-play knowledge graph contrastive learning method named KGE-SymCL by mining the symmetrical structure information in knowledge graphs.

We analyze drawbacks of the exising deep graph clustering methods and scale deep graph clustering to large-scale graphs. The proposed shrink and dilation loss functions optimize clustering distribution adversarially, allowing batch training without performance dropping.

We propose to replace the complicated and consuming graph data augmentations by designing the parameter un-shared siamese encoders and perurbing the node embeddings.

The negative transfer in molecular graph pre-training are analyzed. To alleviate this issue, we first enlarge the atom vocabulary size and then develop two novel pre-training strategies at node and graph level.

We propose Hard Sample Aware Network (HSAN) to mine both the hard positive samples and hard negative samples with a comprehensive similarity measure criterion and a general dynamic sample weighing strategy.

Contrastive deep Graph Clustering network (CCGC) is proposed by mining the intrinsic supervision information in the high-confidence clustering results.

We design a new contrastive learning framework with subgraph-subgraph contrast for the graph anomaly detection.

Deep graph clustering, which aims to group the nodes in graph into disjoint clusters, has become a new hot research spot. This paper summarizes the taxonomy, challenge, and application of deep graph clustering. We hope this work will serve as a quick guide and help researchers to overcome the challenges in this field.

We mathematically disassemble the noise within kernel partition into N-noise and C-noise, and propose a parameter-free model termed MKCDNM, to minimize the dual noise in late fusion framework.

We design a graph-oriented imputation framework called Initializing Then Refining (ITR), which first employs the structure information as the initial imputation, and then adaptively refine the imputed latent variables.

We propose a self-supervised deep graph clustering method termed Dual Correlation Reduction Network (DCRN) to address the representation collapse issue by reducing information correlation in both sample and feature levels.