Automated Black-box Prompt Engineering for Personalized Text-to-Image Generation
This addresses the need for efficient and accessible control over text-to-image generation for users, though it is incremental as it builds on existing automated prompt methods.
The paper tackles the problem of labor-intensive prompt engineering for text-to-image models by introducing PRISM, an algorithm that automatically generates human-interpretable and transferable prompts using black-box access, achieving effective results across models like Stable Diffusion, DALL-E, and Midjourney.
Prompt engineering is an effective but labor-intensive way to control text-to-image (T2I) generative models. Its time-intensive nature and complexity have spurred the development of algorithms for automated prompt generation. However, these methods often struggle with transferability across T2I models, require white-box access to the underlying model, or produce non-intuitive prompts. In this work, we introduce PRISM, an algorithm that automatically produces human-interpretable and transferable prompts that can effectively generate desired concepts given only black-box access to T2I models. Inspired by large language model (LLM) jailbreaking, PRISM leverages the in-context learning ability of LLMs to iteratively refine the candidate prompt distribution built upon the reference images. Our experiments demonstrate the versatility and effectiveness of PRISM in generating accurate prompts for objects, styles, and images across multiple T2I models, including Stable Diffusion, DALL-E, and Midjourney.