Guanfeng Liu

Contrastive Flow Matching for Collaborative Filtering

Wangyu Jin, Jiansheng Qian, Wenwen Xia, Hongliang He et autres

cn, au (code pays fourni par la source)

RAIDS: Rethinking Data Systems as Responsible Intelligent Infrastructure

Zhengyi Yang, Wenke Yang, Guanfeng Liu, Lu Qin

Data systems are evolving from information infrastructure into decision infrastructure. Yet responsibility mechanisms have not kept pace: an output can be accurate or efficient while still lacking sufficient support, satisfied constraints, and actionability for responsible use. We propose RAIDS (Responsible and Intelligent …

RAIDS: Rethinking Data Systems as Responsible Intelligent Infrastructure

Zhengyi Yang, Wenke Yang, Guanfeng Liu, Lu Qin

Data systems are evolving from information infrastructure into decision infrastructure. Yet responsibility mechanisms have not kept pace: an output can be accurate or efficient while still lacking sufficient support, satisfied constraints, and actionability for responsible use. We propose RAIDS (Responsible and Intelligent …

au (code pays fourni par la source)

How Do Graph Signals Affect Recommendation: Unveiling the Mystery of Low and High-Frequency Graph Signals

Feng Liu, Hao Cang, Jiaqing Fan, Yongjing Hao et autres

Spectral graph neural networks (GNNs) are highly effective in modeling graph signals, with their success in recommendation often attributed to low-pass filtering. However, recent studies highlight the importance of high-frequency signals. The role of low-frequency and high-frequency graph signals in recommendation remains …

cn, au (code pays fourni par la source)

Re-understanding Graph Unlearning through Memorization

Pengfei Ding, Yan Wang, Guanfeng Liu

Graph unlearning (GU), which removes nodes, edges, or features from trained graph neural networks (GNNs), is crucial in Web applications where graph data may contain sensitive, mislabeled, or malicious information. However, existing GU methods lack a clear understanding of the key factors …

au (code pays fourni par la source)

Frequency-Corrupt Based Graph Self-Supervised Learning

Haojie Li, Mengjiao Zhang, Guanfeng Liu, Qiang Hu et autres

Graph self-supervised learning (GSSL) alleviates the graph data labeling bottleneck without supervision, enabling wide application in domains like recommendation systems and social network analysis. High-frequency signals are valuable in GSSL for capturing local structural preferences, thereby enriching graph representations and boosting model …

cn, au (code pays fourni par la source)

MARCH: Multi-Teacher Contrastive Hypergraph Distillation

Rongwei Xu, Zitai Qiu, Pengfei Ding, Jia Wu et autres

Recently, hypergraph knowledge distillation has been proposed to alleviate the high computational cost of Hypergraph Neural Networks (HGNNs) when modeling high-order relationships in Web-related graph tasks. Its effectiveness primarily depends on the quality of knowledge transferred from the teacher and the representation …

au (code pays fourni par la source)

LHG: LLM-enhanced and Heterogeneous Graph-induced for Unsupervised Social Event Detection

Zitai Qiu, Rongwei Xu, Congbo Ma, Shan Xue et autres

au, ae, cn, hk (code pays fourni par la source)

Adaptive and Reinforcement-Guided Contrastive Hypergraph Distillation

Rongwei Xu, Zitai Qiu, Pengfei Ding, Guanfeng Liu

Hypergraph-based distillation methods have been proposed to mitigate the high computational cost of Hypergraph Neural Networks (HGNNs) in modeling high-order relationships. However, most existing methods use static and uniform distillation strategies for all nodes and hyperedges, ignoring their individual characteristics. In addition, …

au (code pays fourni par la source)

Structure-attribute aligned graph neural network for recommendation

Zhenduo Qi, Minying Fang, Haojie Li, Feng Jiang et autres

cn, au (code pays fourni par la source)

Wavelet Enhanced Adaptive Frequency Filter for Sequential Recommendation

Huayang Xu, Guanfeng Liu, Junhua Fang, Lei Zhao et autres

Sequential recommendation has garnered significant attention for its ability to capture dynamic preferences by mining users' historical interaction data. Given that users' complex and intertwined periodic preferences are difficult to disentangle in the time domain, recent research is exploring frequency domain analysis …

Intent-Based Trust Evaluation

Rongwei Xu, Guanfeng Liu, Yan Wang, Xuyun Zhang et autres

Trust relationships play a crucial role in various domains, such as social spam detection, retweet behavior analytics, and recommendation systems. Trust is often implicit and difficult to observe directly in the real world, as it is driven by people's underlying intentions and …

au, cn, hk (code pays fourni par la source)