TripletCLIP: Improving Compositional Reasoning of CLIP via Vision-Language Negatives. Patel, M., Kusumba, N. S. A., Cheng, S., Kim, C., Gokhale, T., Baral, C., & Yang, Y. NeuRIPs 2024, 2024.
TripletCLIP: Improving Compositional Reasoning of CLIP via Vision-Language Negatives [link]Paper  bibtex   
@article{patel2024tripletclip,
  title={TripletCLIP: Improving Compositional Reasoning of CLIP via Vision-Language Negatives},
  author={Patel, Maitreya and Kusumba, Naga Sai Abhiram and Cheng, Sheng and Kim, Changhoon and Gokhale, Tejas and Baral, Chitta and Yang, Yezhou},
  url="https://neurips.cc/virtual/2024/poster/94621",
  publisher="NeuRIPs 2024",
  year={2024}
}

Downloads: 0