Benchmark
Pick-a-Pic
A human preference dataset and evaluation benchmark for text-to-image generation. It contains pairwise preference annotations collected from users on generated images and is widely used for training and evaluating image reward models.
Timeline (last 90 days)
Oct 5
On the Pick-a-Pic dataset using DreamSim as the diversity metric, with FLUX.1-dev as the base model and HPSv3 as the reward model, SatisDive improved the worst candidate reward by up to 0.43 compared to FK steering.
Unverified50%