Ye, Y., Le, T. and Lee, D. (2025) “NoisyHate: Mining Online Human-Written Perturbations for Realistic Robustness Benchmarking of Content Moderation Models”, Proceedings of the International AAAI Conference on Web and Social Media, 19(1), pp. 2603–2612. doi: 10.1609/icwsm.v19i1.35961.