Following the convention of RS, existing methods exploit unique user representation in their model design.

Following the convention of RS, existing methods exploit unique user representation in their model design.

We present the T emporal Graph Benchmark (TGB), a collection of challenging and diverse benchmark datasets for realistic, reproducible, and robust evaluation of machine learning models on temporal graphs.

How to construct contrastive views for hypergraphs via augmentations? We provide the solutions in two folds.