jinxixiang

extending VLMO with MIM (Masked Image Modeling) loss

10

Thank you for sharing the source code of VLMO recently. We took a stab and pretrained a large (1024 hidden dim) multiway transformer with mim loss, mlm loss, and contrastive...

We offer a DeepSpeed training code for rapid prototyping. Check it out if you are interested.

4

https://github.com/jinxixiang/magic_animate_unofficial

Watchdog caught collective operation timeout: WorkNCCL

3

Thank you for your contribution to this seminal work! I tried to pretrain a model from scratch, and I prepare the dataset following the examples you provided. Specifically, I collect...

fvd performance on MSR-VTT

2

Thank you for presenting such an exciting work. Congratulations! I have a question regarding Table A3. Could you please provide more details on how the FVD is calculated? As this...

please do not use BiomedCLIP for ARCH dataset

2

Dear Author, The ARCH dataset is divided into two subsets: the **books_set** and the **pubmed_set**. I have noticed that the **pubmed_set** appears to overlap with BioMedCLip, which sources from PubMed...