Releases TencentGR-1M and TencentGR-10M datasets with baselines for all-modality generative recommendation in advertising, including weighted evaluation for conversions.
Title resolution pending
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
roles
method 1polarities
use method 1representative citing papers
CID-TKG combines historical invariance and evolutionary dynamics graphs with contrastive alignment of view-specific relation representations to reach state-of-the-art performance on temporal knowledge graph extrapolation.
Graph Retention Networks extend retention to dynamic graphs to enable parallelizable training, O(1) inference, and chunkwise long-term training while delivering competitive performance with major efficiency gains.
Tempest is a GPU engine that uses dual-index storage and a hierarchical scheduler to deliver high-throughput streaming temporal random walks on billion-edge graphs while enforcing causal order.
PhishEye uses temporal graph contrastive learning on heterogeneous Ethereum transaction graphs for self-supervised phishing detection, achieving F1 scores of 87.23% for transactions and 94.19% for accounts while identifying 1,803 new phishing addresses in real-world deployment that prevented over 2B
citing papers explorer
-
Tencent Advertising Algorithm Challenge 2025: All-Modality Generative Recommendation
Releases TencentGR-1M and TencentGR-10M datasets with baselines for all-modality generative recommendation in advertising, including weighted evaluation for conversions.
-
CID-TKG: Collaborative Historical Invariance and Evolutionary Dynamics Learning for Temporal Knowledge Graph Reasoning
CID-TKG combines historical invariance and evolutionary dynamics graphs with contrastive alignment of view-specific relation representations to reach state-of-the-art performance on temporal knowledge graph extrapolation.
-
Graph Retention Networks for Dynamic Graphs
Graph Retention Networks extend retention to dynamic graphs to enable parallelizable training, O(1) inference, and chunkwise long-term training while delivering competitive performance with major efficiency gains.
-
A GPU Accelerated Temporal Window-Based Random Walk Sampler
Tempest is a GPU engine that uses dual-index storage and a hierarchical scheduler to deliver high-throughput streaming temporal random walks on billion-edge graphs while enforcing causal order.
-
Phishing Detection in Ethereum via Temporal Graph Contrastive Learning
PhishEye uses temporal graph contrastive learning on heterogeneous Ethereum transaction graphs for self-supervised phishing detection, achieving F1 scores of 87.23% for transactions and 94.19% for accounts while identifying 1,803 new phishing addresses in real-world deployment that prevented over 2B