How to understand the accuracy calculation method

THUDM / ImageReward

[NeurIPS 2023] ImageReward: Learning and Evaluating Human Preferences for Text-to-image Generation

Apache License 2.0

1.19k stars 65 forks source link

Open learn01one opened 1 year ago

learn01one commented 1 year ago

Hello, I don’t quite understand the accuracy calculation process. Can you please explain it? Thank you.

xujz18 commented 1 year ago

item_base indicates image rankings of human annotation, while item is the score given by reward model.