Bi-directional Contrastive Distillation for Multi-behavior Recommendation

Article Computer Science, Artificial Intelligence

Latent Structure Mining With Contrastive Modality Fusion for Multimedia Recommendation

Jinghao Zhang et al.

Summary: Multimedia recommendation has become a popular topic in recent years due to the prevalence of multimedia contents on the modern Web. However, previous studies have limitations in modeling item relationships and fusing multiple modalities effectively. To address these issues, this study proposes the MICRO model, which includes a modality-aware structure learning module and a multimodal contrastive framework. Experimental results on real-world datasets demonstrate the superiority of the MICRO model over existing methods.

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING (2023)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence

Contrastive Meta Learning with Behavior Multiplicity for Recommendation

Wei Wei et al.

Summary: A well-informed recommendation framework is important for both user satisfaction and online platform revenue. Existing recommendation models fail to handle multi-typed user behavior data and personalized multi-behavior patterns. To address these challenges, the contrastive meta learning (CML) model is proposed, which utilizes contrastive learning and a meta network to capture behavior multiplicity in recommendation.

WSDM'22: PROCEEDINGS OF THE FIFTEENTH ACM INTERNATIONAL CONFERENCE ON WEB SEARCH AND DATA MINING (2022)

添加到收藏夹

Proceedings Paper Computer Science, Information Systems

Graph Meta Network for Multi-Behavior Recommendation

Lianghao Xia et al.

Summary: Modern recommender systems embed users and items into low-dimensional latent representations based on observed interactions. Users often exhibit various intents driving them to interact with items of multiple behavior types, but existing approaches frequently ignore the diversity of user behaviors, making it challenging to capture relational structures across different behavior types.

SIGIR '21 - PROCEEDINGS OF THE 44TH INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL (2021)

添加到收藏夹

Proceedings Paper Computer Science, Information Systems