创造，不要复制！创造性生成的推动能量扩散

ProCreate, Dont Reproduce! Propulsive Energy Diffusion for Creative Generation

August 5, 2024

作者: Jack Lu, Ryan Teehan, Mengye Ren

cs.AI

摘要

本文提出了ProCreate，这是一种简单易实现的方法，用于提高基于扩散的图像生成模型的样本多样性和创造力，并防止训练数据的复制。ProCreate作用于一组参考图像，并在生成过程中积极推动生成的图像嵌入远离参考嵌入。我们提出了FSCG-8（Few-Shot Creative Generation 8），这是一个少样本创造性生成数据集，涵盖了八个不同类别，包括不同概念、风格和设置，其中ProCreate实现了最高的样本多样性和保真度。此外，我们展示了ProCreate在使用训练文本提示进行大规模评估时有效防止复制训练数据。代码和FSCG-8可在https://github.com/Agentic-Learning-AI-Lab/procreate-diffusion-public获取。项目页面位于https://procreate-diffusion.github.io。

English

In this paper, we propose ProCreate, a simple and easy-to-implement method to improve sample diversity and creativity of diffusion-based image generative models and to prevent training data reproduction. ProCreate operates on a set of reference images and actively propels the generated image embedding away from the reference embeddings during the generation process. We propose FSCG-8 (Few-Shot Creative Generation 8), a few-shot creative generation dataset on eight different categories -- encompassing different concepts, styles, and settings -- in which ProCreate achieves the highest sample diversity and fidelity. Furthermore, we show that ProCreate is effective at preventing replicating training data in a large-scale evaluation using training text prompts. Code and FSCG-8 are available at https://github.com/Agentic-Learning-AI-Lab/procreate-diffusion-public. The project page is available at https://procreate-diffusion.github.io.