no code implementations • 12 Jan 2024 • Jiangming Shi, Xiangbo Yin, Yeyun Chen, Yachao Zhang, Zhizhong Zhang, Yuan Xie, Yanyun Qu
To associate cross-modality clustered pseudo-labels, we design a Multi-Memory Learning and Matching (MMLM) module, ensuring that optimization explicitly focuses on the nuances of individual perspectives and establishes reliable cross-modality correspondences.