search

TIGER-AI-Lab / Mantis

Official code for Paper "Mantis: Multi-Image Instruction Tuning"
https://tiger-ai-lab.github.io/Mantis/
Apache License 2.0
176 stars 15 forks source link

Confusion about Eval and Instruct in README #21

Open emanuelevivoli opened 1 month ago

emanuelevivoli commented 1 month ago

Hello! I'm a big fan of your Mantis paper, I really like it! (and thanks for this repo!)

I have a simple question, to clarify the reproducibility. In the README there is written:

Intermediate Checkpoints The following intermediate checkpoints after pre-training the multi-modal projectors are also available for experiment reproducibility (Please note the following checkpoints still needs further fine-tuning on Mantis-Eval to be intelligent. They are not working models.): TIGER-Lab/Mantis-8B-clip-llama3-pretraind TIGER-Lab/Mantis-8B-siglip-llama3-pretraind

But I assume there is a mistake and it should be Mantis-Instruct, right?

image
jdf-prog commented 1 day ago

Yeah, it's a typo. Thanks for pointing out and sorry for the late response. It should be Mantis-Instruct instead of Mantis-Eval. I now have fixed this typo!