Cyberbullying classifiers are sensitive to model-agnostic perturbations
C Emmery, A Kádár, G Chrupała… - arXiv preprint arXiv …, 2022 - arxiv.org
A limited amount of studies investigates the role of model-agnostic adversarial behavior in
toxic content classification. As toxicity classifiers predominantly rely on lexical …
toxic content classification. As toxicity classifiers predominantly rely on lexical …
User-Centered Security in Natural Language Processing
C Emmery - arXiv preprint arXiv:2301.04230, 2023 - arxiv.org
This dissertation proposes a framework of user-centered security in Natural Language
Processing (NLP), and demonstrates how it can improve the accessibility of related …
Processing (NLP), and demonstrates how it can improve the accessibility of related …