PLADIS: Pushing the Limits of Attention in Diffusion Models at Inference Time by Leveraging Sparsity
This provides a more efficient and universally applicable solution for enhancing pre-trained diffusion models, particularly benefiting users of guidance-distilled models.
The authors tackled the problem of improving text-to-image diffusion models at inference time without extra training or neural function evaluations, achieving notable improvements in text alignment and human preference.
Diffusion models have shown impressive results in generating high-quality conditional samples using guidance techniques such as Classifier-Free Guidance (CFG). However, existing methods often require additional training or neural function evaluations (NFEs), making them incompatible with guidance-distilled models. Also, they rely on heuristic approaches that need identifying target layers. In this work, we propose a novel and efficient method, termed PLADIS, which boosts pre-trained models (U-Net/Transformer) by leveraging sparse attention. Specifically, we extrapolate query-key correlations using softmax and its sparse counterpart in the cross-attention layer during inference, without requiring extra training or NFEs. By leveraging the noise robustness of sparse attention, our PLADIS unleashes the latent potential of text-to-image diffusion models, enabling them to excel in areas where they once struggled with newfound effectiveness. It integrates seamlessly with guidance techniques, including guidance-distilled models. Extensive experiments show notable improvements in text alignment and human preference, offering a highly efficient and universally applicable solution. See Our project page : https://cubeyoung.github.io/pladis-proejct/