X
Xu Sun
Researcher at The University of Nottingham Ningbo China
Publications - 329
Citations - 6766
Xu Sun is an academic researcher from The University of Nottingham Ningbo China. The author has contributed to research in topics: Computer science & Deep learning. The author has an hindex of 41, co-authored 296 publications receiving 4581 citations. Previous affiliations of Xu Sun include Dalian Medical University & University of Tokyo.
Papers
More filters
Posted Content
Measuring and Relieving the Over-smoothing Problem for Graph Neural Networks from the Topological View
TL;DR: Two methods to alleviate the over-smoothing issue of GNNs are proposed: MADReg which adds a MADGap-based regularizer to the training objective; AdaEdge which optimizes the graph topology based on the model predictions.
Journal ArticleDOI
Measuring and Relieving the Over-Smoothing Problem for Graph Neural Networks from the Topological View
TL;DR: Wang et al. as discussed by the authors proposed two methods to alleviate the over-smoothing issue from the topological view: (1) MADReg which adds a MADGap-based regularizer to the training objective; (2) AdaEdge which optimizes the graph topology based on the model predictions.
Proceedings Article
SGM: Sequence Generation Model for Multi-label Classification
TL;DR: This paper proposes to view the multi-label classification task as a sequence generation problem, and apply a sequencegeneration model with a novel decoder structure to solve it, and shows that the proposed methods outperform previous work by a substantial margin.
Proceedings ArticleDOI
Unpaired Sentiment-to-Sentiment Translation: A Cycled Reinforcement Learning Approach
TL;DR: A cycled reinforcement learning method that enables training on unpaired data by collaboration between a neutralization module and an emotionalization module that significantly outperforms the state-of-the-art systems.
Proceedings ArticleDOI
A Dual Reinforcement Learning Framework for Unsupervised Text Style Transfer
TL;DR: This paper proposed a dual RL framework to directly transfer the style of the text via a one-step mapping model, without any separation of content and style, and two rewards are designed based on such a dual structure to reflect the style accuracy and content preservation, respectively.