About me

Hey! I’m Luiza. I’m a Research Scholar at Cohere For AI, where I work with Beyza Ermis, Sara Hooker, and Patrick Lewis. Also, I’m a Master’s student at Unicamp, Brazil advised by the amazing professors Paula Dornhofer and Eduardo Valle. Currently, I’m exploring toxicity mitigation of language models with retrieval-augmentation techniques. I’m also broadly interested in AI Ethics and Safety, model robustness, interpretability, and data-centric AI.

News

• 2023.01: I’m starting as a Research Scholar at Cohere For AI!! I’ll be working on toxicity mitigation of language models.