Don't be a Fool: Pooling Strategies in Offensive Language Detection from User-Intended Adversarial Attacks

AAMLRe-assign community