I am a Research Scientist at ByteDance, London, where I am working on LLM reasoning and multi-agent collaboration. My broader research interests span the robustness and safety of AI systems, with a focus on uncertainty quantification, causal inference, algorithmic fairness, and interpretability.
Prior to this, I completed my PhD in machine learning at Oxford, under the supervision of Rob Cornish, Arnaud Doucet and Yee Whye Teh, and funded by Google DeepMind. You can find more details in my PhD thesis.
I am also a part of the organising committee for ELLIS RobustML workshops. In 2024, we organised two robustness-related workshops with the aim of bringing together theoretical robustness research with real-world applications. For updates regarding upcoming workshops, join our mailing group.
Contact
Feel free to reach out to me at faaiz.taufiq (at) bytedance.com