ImageReward
A reward model for text-to-image tasks, trained on human preference feedback data. It evaluates the alignment between generated images and text prompts as well as overall image quality.
Timeline (last 90 days)
With SANA-1.6B as the base model and ImageReward as the reward model, SatisDive improved the worst candidate reward by up to 0.70 compared to FK steering.
Both HPSv3 and ImageReward are human preference reward models trained on large-scale human-annotated image preference datasets to predict human satisfaction scores for generated images.
All Facts (2)
With SANA-1.6B as the base model and ImageReward as the reward model, SatisDive improved the worst candidate reward by up to 0.70 compared to FK steering.
50%UnverifiedBoth HPSv3 and ImageReward are human preference reward models trained on large-scale human-annotated image preference datasets to predict human satisfaction scores for generated images.
50%