FLUX.1-dev
A high-quality text-to-image diffusion model released by Black Forest Labs. The developer version (dev) supports both research and commercial application scenarios.
Timeline (last 90 days)
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.
SatisDive requires no training and intervenes only at inference time, possessing plug-and-play characteristics; the paper has been validated across multiple combinations including FLUX.1-dev, SANA-1.6B, HPSv3, and ImageReward.
All Facts (2)
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.
50%UnverifiedSatisDive requires no training and intervenes only at inference time, possessing plug-and-play characteristics; the paper has been validated across multiple combinations including FLUX.1-dev, SANA-1.6B, HPSv3, and ImageReward.
50%