自动生成负面提示，让AI更准确理解文本描述。
Guiding What Not to Generate: Automated Negative Prompting for Text-Image Alignment
- 通过分析注意力模式，自动发现应避免生成的内容。
- 在GenEval++上达到0.571分数，优于基线0.371。
- 无需额外图像生成，适合想提升生成精度的用户。
尽管文本到图像生成已取得显著进展，但在包含复杂构图或想象元素的提示下实现精确对齐仍具挑战。为此，我们提出图像修正用负提示（NPC），一种自动化流程，通过识别并应用抑制非预期内容的负提示来改善对齐。首先分析交叉注意力模式，揭示为何目标性负提示（直接关联对齐错误）与非目标性负提示（与提示无关但出现在生成图像中）均能提升对齐效果。NPC采用验证器-标题生成器-提议者框架生成候选提示，并通过显著文本空间得分进行排序，无需额外图像合成即可有效选择。在GenEval++和Imagine-Bench上，NPC超越强基线，在GenEval++上达0.571分（基线0.371），在Imagine-Bench上表现最佳。通过指导‘不生成什么’，NPC为扩散模型提供了原则性、全自动的更强文本图像对齐路径。代码已开源：https://github.com/wiarae/NPC。
原文摘要 · Abstract (English)
Despite substantial progress in text-to-image generation, achieving precise text-image alignment remains challenging, particularly for prompts with rich compositional structure or imaginative elements. To address this, we introduce Negative Prompting for Image Correction (NPC), an automated pipeline that improves alignment by identifying and applying negative prompts that suppress unintended content. We begin by analyzing cross-attention patterns to explain why both targeted negatives-those directly tied to the prompt's alignment error-and untargeted negatives-tokens unrelated to the prompt but present in the generated image-can enhance alignment. To discover useful negatives, NPC generates candidate prompts using a verifier-captioner-proposer framework and ranks them with a salient text-space score, enabling effective selection without requiring additional image synthesis. On GenEval++ and Imagine-Bench, NPC outperforms strong baselines, achieving 0.571 vs. 0.371 on GenEval++ and the best overall performance on Imagine-Bench. By guiding what not to generate, NPC provides a principled, fully automated route to stronger text-image alignment in diffusion models. Code is released at https://github.com/wiarae/NPC.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。