photo of fireworks

A team led by senior BMI PhD students, Gautam Machiraju and Alex Derry, is excited to announce a new paper accepted for publication in the 2024 International Conference on Machine Learning (ICML) in Vienna, Austria. This work advances new modes of inference for Foundation Models, allowing them to efficiently and accurately identify class-specific features in various data modalities (sequences, images, graphs). This work was advised by multiple Stanford-based faculty including Parag Mallick, Christopher Ré, James Zou, and Russ Altman. Congrats to the team!