search

X-PLUG / mPLUG-Owl

mPLUG-Owl: The Powerful Multi-modal Large Language Model Family
https://www.modelscope.cn/studios/damo/mPLUG-Owl
MIT License
2.25k stars 171 forks source link

Did you use the full LAION-400M and COYO-700M dataset for pre-training, or just sampled subsets. What's your total amount of image-text pairs for pre-training? #162

Closed linserSnow closed 11 months ago

MAGAer13 commented 11 months ago

In the appendix, we only utilize about 400 millions image-text pairs from LAION-400M and COYO-700M. We did not use the whole datasets.