Cross-User Generalization and Personalization for Multimodal Smart-Glass Data
A research project at USI studying cross-user domain shift and low-label personalization for multimodal smart-glass sensor data.
Self-Supervised Learning
Accurate prediction of antibody function and structure using bio-inspired antibody language model
We developed BALM, a bio-inspired antibody language model with antibody-specific inductive biases for accurate prediction of antibody function and structure.