Tingyang Xu

Unsupervised Adversarial Graph Alignment with Graph Embedding

2019 · arXiv (Cornell University)

Graph alignment, also known as network alignment, is a fundamental task in social network analysis. Many recent works have relied on partially labeled cross-graph node correspondences, i.e., anchor links. However, due to the privacy and …

A Restricted Black-Box Adversarial Framework Towards Attacking Graph Embedding Models

2020 · Proceedings of the AAAI Conference on Artificial Intelligence

With the great success of graph embedding model on both academic and industry area, the robustness of graph embedding against adversarial attack inevitably becomes a central problem in graph learning domain. Regardless of the fruitful …

On Self-Distilling Graph Neural Network

2020 · arXiv (Cornell University)

Recently, the teacher-student knowledge distillation framework has demonstrated its potential in training Graph Neural Networks (GNNs). However, due to the difficulty of training over-parameterized GNN models, one may not easily obtain a satisfactory teacher model …

Recognizing Predictive Substructures With Subgraph Information Bottleneck

2021 · IEEE Transactions on Pattern Analysis and Machine Intelligence

The emergence of Graph Convolutional Network (GCN) has greatly boosted the progress of graph learning. However, two disturbing factors, noise and redundancy in graph data, and lack of interpretation for prediction results, impede further development …

Towards the Explanation of Graph Neural Networks in Digital Pathology with Information Flows

2021 · arXiv (Cornell University)

As Graph Neural Networks (GNNs) are widely adopted in digital pathology, there is increasing attention to developing explanation models (explainers) of GNNs for improved transparency in clinical decisions. Existing explainers discover an explanatory subgraph relevant …

Large Language-Geometry Model: When LLM meets Equivariance

2025 · arXiv (Cornell University)

Accurately predicting 3D structures and dynamics of physical systems is crucial in scientific applications. Existing approaches that rely on geometric Graph Neural Networks (GNNs) effectively enforce $\mathrm{E}(3)$-equivariance, but they often fall in leveraging extensive broader …

Graph Representation Learning via Graphical Mutual Information Maximization

2020

The richness in the content of various information networks such as social networks and communication networks provides the unprecedented potential for learning high-quality expressive representations without external supervision. This paper investigates how to preserve and …