Weights for Semantic-Consistent Dual-Image Vision-Language Alignment via Modality Routing for 3D PET/CT (MICCAI 2026).
Code: https://github.com/Linn60/PETCT-DiVLA
