If you make use of the code/experiment or TBGAT algorithm in your work, please cite our paper (Bibtex below). @InProceedings{zhanglearning2024, title = {Learning Topological Representations with ...
Phase 1: Content encoder (pretrained ResNet50 backbone) + style encoder trained with contrastive loss (NT-Xent) to disentangle content from style. Phase 2: Joint training with a decoder for ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results