Scholarlink

<p>Text-to-image diffusion models have demonstrated unprecedented capabilities for flexible and realistic image synthesis. Nevertheless, these models rely on a time-consuming sampling procedure, which has motivated attempts to reduce their latency. When improving efficiency, researchers often use the original diffusion model to train an additional network designed specifically for fast image generation. In contrast, our approach seeks to reduce latency directly, without any retraining, fine-tuning, or knowledge distillation. In particular, we find the repeated calculation of attention maps to be costly yet redundant, and instead suggest reusing them during sampling. Our specific reuse strategies are based on ODE theory, which implies that the later a map is reused, the smaller the distortion in the final image. We empirically compare our reuse strategies with few-step sampling procedures of comparable latency, finding that reuse generates images that are closer to those produced by the original high-latency diffusion model.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Fast Sampling Through The Reuse Of Attention Maps In Diffusion Models

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年6月11日

查看全文
  • Rosco Hunter,
  • Łukasz Dudziak,
  • Mohamed S. Abdelfattah,
  • Abhinav Mehrotra,
  • Sourav Bhattacharya,
  • Hongkai Wen

摘要

Text-to-image diffusion models have demonstrated unprecedented capabilities for flexible and realistic image synthesis. Nevertheless, these models rely on a time-consuming sampling procedure, which has motivated attempts to reduce their latency. When improving efficiency, researchers often use the original diffusion model to train an additional network designed specifically for fast image generation. In contrast, our approach seeks to reduce latency directly, without any retraining, fine-tuning, or knowledge distillation. In particular, we find the repeated calculation of attention maps to be costly yet redundant, and instead suggest reusing them during sampling. Our specific reuse strategies are based on ODE theory, which implies that the later a map is reused, the smaller the distortion in the final image. We empirically compare our reuse strategies with few-step sampling procedures of comparable latency, finding that reuse generates images that are closer to those produced by the original high-latency diffusion model.