Yuhan Zhu

Thank you for your great work. ASL adopts 'clamp' to prevent 'inf' in the loss calculation. Actually, 'clamp' cannot back-propagate the gradients, because it is not derivable, i.e., ASL ignores...

Hi Jindong, Thanks for the great survey! I've submitted a PR to add three 2024 papers: - "AWT: Transferring Vision-Language Models via Augmentation, Weighting, and Transportation" - Accepted at NeurIPS...

Request to add new papers: AWT, ProVP, Self-TPT

1

Hi Zheng, Thank you for your fantastic repository! I'm wondering if you could consider including my recent papers: - [AWT](https://arxiv.org/abs/2407.04603) [Code](https://github.com/MCG-NJU/AWT) (NeurIPS 2024): This paper discusses transferring VLMs to downstream...