• Publications
  • Influence
Network Representation Learning with Rich Text Information
TLDR
By proving that DeepWalk, a state-of-the-art network representation method, is actually equivalent to matrix factorization (MF), this work proposes text-associated DeepWalk (TADW), which incorporates text features of vertices into network representation learning under the framework of Matrix factorization.
GEAR: Graph-based Evidence Aggregating and Reasoning for Fact Verification
TLDR
A graph-based evidence aggregating and reasoning (GEAR) framework which enables information to transfer on a fully-connected evidence graph and then utilizes different aggregators to collect multi-evidence information is proposed.
Fast Network Embedding Enhancement via High Order Proximity Approximation
TLDR
Most existing NRL methods are summarized into a unified two-step framework, including proximity matrix construction and dimension reduction, and Network Embedding Update (NEU) algorithm is proposed which implicitly approximates higher order proximities with theoretical approximation bound.
A Neural Network Approach to Jointly Modeling Social Networks and Mobile Trajectories
TLDR
A novel neural network model is presented that can jointly model both social networks and mobile trajectories and the effectiveness of this model is demonstrated when either network structure or trajectory data is sparse.
Adaptive Graph Encoder for Attributed Graph Embedding
TLDR
Experimental results show that AGE consistently outperforms state-of-the-art graph embedding methods considerably on node clustering and link prediction tasks, and the proposed Adaptive Graph Encoder employs an adaptive encoder that iteratively strengthens the filtered features for better node embeddings.
CED: Credible Early Detection of Social Media Rumors
TLDR
A novel early rumor detection model, Credible Early Detection (CED), regarding all reposts to a rumor candidate as a sequence, which can remarkably reduce the time span for prediction by more than 85 percent, with better accuracy performance than all state-of-the-art baselines.
Zero-Shot Cross-Lingual Neural Headline Generation
TLDR
This work proposes to deal with the cross-lingual neural headline generation (CNHG) under the zero-shot scenario, and lets a parameterized CNHG model (student model) mimic the output of a pretrained translation or headline generation model (teacher model).
Extract the Knowledge of Graph Neural Networks and Go Beyond it: An Effective Knowledge Distillation Framework
TLDR
This paper proposes a framework based on knowledge distillation that extracts the knowledge of an arbitrary learned GNN model (teacher model), and injects it into a well-designed student model, which is designed as a trainable combination of parameterized label propagation and feature transformation modules.
...
...