FABRIC: Personalizing Diffusion Models with Iterative Feedback
This work addresses the need for personalized content creation and customization in visual generation, offering an incremental improvement by adapting existing diffusion models with feedback mechanisms.
The study tackled the problem of integrating iterative human feedback into diffusion-based text-to-image models to enhance output quality, proposing FABRIC, a training-free method that improved generation results over multiple feedback rounds.
In an era where visual content generation is increasingly driven by machine learning, the integration of human feedback into generative models presents significant opportunities for enhancing user experience and output quality. This study explores strategies for incorporating iterative human feedback into the generative process of diffusion-based text-to-image models. We propose FABRIC, a training-free approach applicable to a wide range of popular diffusion models, which exploits the self-attention layer present in the most widely used architectures to condition the diffusion process on a set of feedback images. To ensure a rigorous assessment of our approach, we introduce a comprehensive evaluation methodology, offering a robust mechanism to quantify the performance of generative visual models that integrate human feedback. We show that generation results improve over multiple rounds of iterative feedback through exhaustive analysis, implicitly optimizing arbitrary user preferences. The potential applications of these findings extend to fields such as personalized content creation and customization.