Multi-modal alignment via hyperbolic geometry

• Main Author: LIU, Suyu
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2024
• Subjects: Multi-modal Learning; Geometric deep learning; Vision-language model; Domain adaptation; Artificial Intelligence and Robotics
• Online Access: https://ink.library.smu.edu.sg/etd_coll/651; https://ink.library.smu.edu.sg/context/etd_coll/article/1649/viewcontent/GPIS_AY2022_MbR_SuyuLiu.pdf
• Institution: Singapore Management University

Strong capabilities of generalization to unseen domains are vital for deep neural networks. While existing methods have shown promising results without source domain access, they mostly rely on models that are extensively pre-trained on source domains or overlook the intricate hierarchical structure...