what is the difference of the second pre-trained model for different tasks?

thunlp / PEVL

Source code for EMNLP 2022 paper “PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models”

MIT License

47 stars 5 forks source link

Open ZHUXUHAN opened 1 year ago

ZHUXUHAN commented 1 year ago

i see there are different pre-trained model for each downstream tasks, what is the difference?