search

yuhui-zh15 / VLMClassifier

Official implementation of "Why are Visually-Grounded Language Models Bad at Image Classification?"
https://yuhui-zh15.github.io/VLMClassifier-Website/
17 stars 1 forks source link

Pretrained weights for fine-tune VLMs? #1

Open ngthanhtin opened 1 week ago

ngthanhtin commented 1 week ago

Dear Yuhui Zhang,

Thank you for your great effort! I found your paper very interesting and informative.

Could you provide us with the pretrained weights for fine-tuned VLMs? It would be a tremendous help for me to continue my research.

Best, Tin

Weibin44 commented 1 week ago

Hi Tin, did you find the code for fine-tuning LlaVA? It seems there are only for clip under train_analysis folder. Thanks for your help!

ngthanhtin commented 1 week ago

Hi @Weibin44 , that's what I am asking for

yuhui-zh15 commented 1 week ago

Thanks for your interest in our work!

We simply use the official repo to fine-tune LLaVA and BLIP2.

We will finish cleaning these fine-tuning codes and releasing model weights in the next month.