The first release of **Graph Robustness Benchmark (GRB)**.
- API based on pure PyTorch, CogDL, and DGL.
- Include five graph [datasets](https://cogdl.ai/grb/datasets) of different scales.
- Support graph injection attacks (e.g., RND, FGSM, PGS, SPEIT, TDGIA).
- Support adversarial defenses (e.g., layer normalization, adversarial training, GNNSVD, GNNGuard).
- Provide [homepage](https://cogdl.ai/grb/home).
- Provide [leaderboards](https://cogdl.ai/grb/leaderboard/) of all datasets.
- Provide basic [documentation](https://grb.readthedocs.io/en/latest/).
- Provide scripts for reproducing results.