Research map / Graph learning / Graph self-supervised learning
GCIB: Graph Contrastive Information Bottleneck for Multi-Behavior Recommendation
Likang Wu, Zihao Chen, Jianxin Zhang, Sangqi Zhu, Yuanyuan Ge, Haipeng Yang, Lei Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add global encoding on heterogeneous multi-behavior graph (ablation) | GCIB · Tmall | HR@10: 0.1101 -> 0.1617 | Improved |
| add global encoding (ablation) | GCIB · Taobao | HR@10: 0.1666 -> 0.1815 | Improved |
| add global encoding (ablation) | GCIB · Yelp | HR@10: 0.0698 -> 0.0746 | Improved |
| add information bottleneck auxiliary denoising (ablation) | GCIB · Tmall | HR@10: 0.1089 -> 0.1617 | Improved |
| add information bottleneck auxiliary denoising (ablation) | GCIB · Taobao | HR@10: 0.1724 -> 0.1815 | Improved |
| add information bottleneck auxiliary denoising (ablation) | GCIB · Yelp | HR@10: 0.0700 -> 0.0746 | Improved |
| add InfoNCE cross-behavior contrastive alignment (ablation) | GCIB · Tmall | HR@10: 0.1523 -> 0.1617 | Improved |
| add InfoNCE cross-behavior contrastive alignment (ablation) | GCIB · Taobao | HR@10: 0.1661 -> 0.1815 | Improved |
| add InfoNCE cross-behavior contrastive alignment (ablation) | GCIB · Yelp | HR@10: 0.0641 -> 0.0746 | Improved |
| add both IB denoising and contrastive alignment (ablation) | GCIB · Tmall | HR@10: 0.0356 -> 0.1617 | Improved |
| add both IB denoising and contrastive alignment (ablation) | GCIB · Taobao | HR@10: 0.0352 -> 0.1815 | Improved |
| add both IB denoising and contrastive alignment (ablation) | GCIB · Yelp | HR@10: 0.0522 -> 0.0746 | Improved |
Closest papers by meaning
- Robust Graph Based Social Recommendation Through Contrastive Multi-View Learning (AAAI 2025 · similarity 0.88)
- Contrastive Graph Structure Learning via Information Bottleneck for Recommendation (NeurIPS 2022 · similarity 0.88)
- Multiple Purchase Chains with Negative Transfer Elimination for Multi-Behavior Recommendation (AAAI 2025 · similarity 0.86)
- DivGCL: A Graph Contrastive Learning Model for Diverse Recommendation (AAAI 2025 · similarity 0.85)
- Graph Disentangled Contrastive Learning with Personalized Transfer for Cross-Domain Recommendation (AAAI 2024 · similarity 0.84)
- Temporal Graph Contrastive Learning for Sequential Recommendation (AAAI 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
