COSMOS is introduced in the paper
COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training. COSMOS is trained in self-supervised learning framework with multi-modal augmentation and cross-attention module. It outperforms CLIP-based models trained on larger datasets in visual perception and contextual understanding tasks. COSMOS also achieves strong performance in downstream tasks including zero-shot image-text retrieval, classification, and semantic segmentation.
1@article{kim2024cosmos,
2 title={COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training},
3 author={Kim, Sanghwan and Xiao, Rui and Georgescu, Mariana-Iuliana and Alaniz, Stephan and Akata, Zeynep},
4 journal={arXiv preprint arXiv:2412.01814},
5 year={2024}
6}
