In text-to-image generation, using negative prompts, which describe undesirable image characteristics, can significantly boost image quality. However, producing good negative prompts is manual and tedious. To address this, we propose NegOpt, a novel method for optimizing negative prompt generation toward enhanced image generation, using supervised fine-tuning and reinforcement learning. Our combined approach results in a substantial increase of 25% in Inception Score compared to other approaches and surpasses ground-truth negative prompts from the test set. Furthermore, with NegOpt we can preferentially optimize the metrics most important to us. Finally, we construct Negative Prompts DB, a dataset of negative prompts.

翻译：在文本到图像生成中，使用描述不良图像特征的负提示可显著提升图像质量。然而，生成优质的负提示需要手动操作且过程繁琐。为解决此问题，我们提出NegOpt——一种通过监督微调与强化学习优化负提示生成的新方法，旨在增强图像生成效果。相比其他方法，我们的联合方法使Inception Score大幅提升25%，甚至超越测试集中的真实负提示。此外，借助NegOpt，我们可优先优化最关注的指标。最后，我们构建了负提示数据集Negative Prompts DB。