2 papers
cs.CV2026
Physics-Aligned Self-Supervised Learning for Scientific Imaging
Bashir Kazimi, Stefan Sandfeld
Data augmentations define the invariances learned by self-supervised learning (SSL). Standard augmentation pipelines were designed for natural images, yet scientific imaging modali…
cs.LG2026
MatBind: A Shared Embedding Space for Multimodal Materials Characterization
Le Yang, Anoop K. Chandran, Jona Ãstreicher +8
Fully characterizing a crystalline material requires integrating heterogeneous data sources -- atomic structures, diffraction patterns, electronic density of states, and natural la…