ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation

04/12/2023
by   Jiazheng Xu, et al.
5

We present ImageReward – the first general-purpose text-to-image human preference reward model – to address various prevalent issues in generative models and align them with human values and preferences. Its training is based on our systematic annotation pipeline that covers both the rating and ranking components, collecting a dataset of 137k expert comparisons to date. In human evaluation, ImageReward outperforms existing scoring methods (e.g., CLIP by 38.6%), making it a promising automatic metric for evaluating and improving text-to-image synthesis. The reward model is publicly available via the package at <https://github.com/THUDM/ImageReward>.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset