Excited to be presenting part of our latest work on machine unlearning, interpretability, and representation learning. Checkout the preprints!
Checkout the preprints!
I presented different lines of research on foundation models and interpretable AI at the Interpretable Deep Learning Reading Group and the University of Zaragoza Seminar Series.
Monthly event exploring how and why neural models develop similar internal representations, and what this means for learning, alignment, and reuse.
Gave an oral presentation on our project on generative vision-language models for Easy Read accessibility. It was great to learn about the many translational projects within the ETH-UN partnership!
Excited to be presenting recent work on interpretable reward modeling for LLMs in RLHF, self-supervised learning to leverage multiple modalities in medical domains, uncertainty estimation in variational networks, and on predictive models in healthcare. Checkout the preprints!
Excited to be presenting part of our latest work on 3D generative models, an information theoretic view of interpretability, and on predictive models in healthcare. Checkout the preprints!
I presented my PhD research as an invited speaker at the Microsoft Cambridge ML Seminar Series, very happy about this opportunity.
Selected candidate to participate in the Spring into Quant Program hosted by G-Research. Grateful for the insightful week learning about Machine Learning in Quant research and engaging workshops, talks and speakers!
Will be working on alignment of LLMs with Prof. Mihaela Van der Shaar.