This project focuses on developing methods for automatically detecting offensive or toxic language in text and transforming it into a more neutral or polite form. The student will explore state-of-the-art natural language processing techniques, including machine learning and large language models, to classify offensive content and generate detoxified alternatives while preserving the original meaning. The project may involve working with existing datasets, evaluating different modeling approaches, and designing a pipeline that combines detection and text rewriting. The outcome will contribute to building safer and more inclusive online communication systems.

References

1. Davidson, T., Warmsley, D., Macy, M. and Weber, I., 2017, May. Automated hate speech detection and the problem of offensive language. In Proceedings of the international AAAI conference on web and social media (Vol. 11, No. 1, pp. 512-515). https://ojs.aaai.org/index.php/ICWSM/article/view/14955
2. Zampieri, M., Malmasi, S., Nakov, P., Rosenthal, S., Farra, N. and Kumar, R., 2019, June. Predicting the type and target of offensive posts in social media. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) (pp. 1415-1420). https://aclanthology.org/N19-1144/
3. Dementieva, D., Protasov, V., Babakov, N., Rizwan, N., Alimova, I., Brun, C., Konovalov, V., Muti, A., Liebeskind, C., Litvak, M. and Nozza, D., 2025. Overview of the multilingual text detoxification task at pan 2025. In CEUR Workshop Proceedings. https://pan.webis.de/clef25/pan25-web/text-detoxification.html?utm_source=chatgpt.com
4. Vanetik, N., Liberov, L., Litvak, M. and Liebeskind, C., 2025, September. Towards Safer Hebrew Communication: A Dataset for Offensive Language Detoxification. In Proceedings of the 15th International Conference on Recent Advances in Natural Language Processing-Natural Language Processing in the Generative AI Era (pp. 1289-1298). https://aclanthology.org/2025.ranlp-1.149/