Výsledky vyhledávání - "Sun Qingqiang"

Report

Out-of-Distribution Detection with Prototypical Outlier Proxy

Autor: Gong, Mingrong, Chen, Chaoqi, Sun, Qingqiang, Wang, Yue, Huang, Hui

Out-of-distribution (OOD) detection is a crucial task for deploying deep learning models in the wild. One of the major challenges is that well-trained deep models tend to perform over-confidence on unseen test data. Recent research attempts to levera

Externí odkaz: http://arxiv.org/abs/2412.16884

Zobrazit plný text záznamu

Report

Single-View Graph Contrastive Learning with Soft Neighborhood Awareness

Autor: Sun, Qingqiang, Chen, Chaoqi, Qiao, Ziyue, Zheng, Xubin, Wang, Kai

Most graph contrastive learning (GCL) methods heavily rely on cross-view contrast, thus facing several concomitant challenges, such as the complexity of designing effective augmentations, the potential for information loss between views, and increase

Externí odkaz: http://arxiv.org/abs/2412.09261

Zobrazit plný text záznamu

Report

PRAGA: Prototype-aware Graph Adaptive Aggregation for Spatial Multi-modal Omics Analysis

Autor: Huang, Xinlei, Ma, Zhiqi, Meng, Dian, Liu, Yanran, Ruan, Shiwei, Sun, Qingqiang, Zheng, Xubin, Qiao, Ziyue

Spatial multi-modal omics technology, highlighted by Nature Methods as an advanced biological technique in 2023, plays a critical role in resolving biological regulatory processes with spatial context. Recently, graph neural networks based on K-neare

Externí odkaz: http://arxiv.org/abs/2409.12728

Zobrazit plný text záznamu

Report

Holistic Memory Diversification for Incremental Learning in Growing Graphs

Autor: Qiao, Ziyue, Xiao, Junren, Sun, Qingqiang, Xiao, Meng, Xiong, Hui

This paper addresses the challenge of incremental learning in growing graphs with increasingly complex tasks. The goal is to continually train a graph model to handle new tasks while retaining its inference ability on previous tasks. Existing methods

Externí odkaz: http://arxiv.org/abs/2406.07413

Zobrazit plný text záznamu

Report

Towards Higher-order Topological Consistency for Unsupervised Network Alignment

Autor: Sun, Qingqiang, Lin, Xuemin, Zhang, Ying, Zhang, Wenjie, Chen, Chaoqi

Network alignment task, which aims to identify corresponding nodes in different networks, is of great significance for many subsequent applications. Without the need for labeled anchor links, unsupervised alignment methods have been attracting more a

Externí odkaz: http://arxiv.org/abs/2208.12463

Zobrazit plný text záznamu