从上下文矛盾的提示词生成图像
Image Generation from Contextually-Contradictory Prompts
把提示词按去噪阶段拆成一系列代理提示词，用 LLM 解析概念冲突并改写，为文本到图像扩散模型提供细粒度的语义控制；这一思路可直接迁移到组合式生成、属性绑定和提示词工程等场景。
原文依据 we propose a stage-aware prompt decomposition framework that guides the denoising process using a sequence of proxy prompts
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论文本到图像扩散模型擅长根据自然语言提示词生成高质量、多样化的图像。然而，当提示词中包含与模型已学习先验相矛盾的概念组合时，它们往往无法生成语义准确的结果。我们将这一失效模式定义为上下文矛盾，即由于训练中学习到的纠缠关联，一个概念会隐式地否定另一个概念。为解决该问题，我们提出一种阶段感知的提示词分解框架，利用一系列代理提示词引导去噪过程。每个代理提示词都被构造为与去噪特定阶段预期出现的语义内容相匹配，同时保证上下文连贯性。为构造这些代理提示词，我们借助大语言模型（LLM）分析目标提示词、识别矛盾，并生成在保留原始意图的同时化解上下文冲突的替代表达。通过将提示词信息与去噪进程对齐，我们的方法能够在存在上下文矛盾的情况下实现细粒度的语义控制和准确的图像生成。在多种具有挑战性的提示词上的实验表明，与文本提示词的对齐程度有显著提升。
英文原文摘要 1257 字符
Text-to-image diffusion models excel at generating high-quality, diverse images from natural language prompts. However, they often fail to produce semantically accurate results when the prompt contains concept combinations that contradict their learned priors. We define this failure mode as contextual contradiction, where one concept implicitly negates another due to entangled associations learned during training. To address this, we propose a stage-aware prompt decomposition framework that guides the denoising process using a sequence of proxy prompts. Each proxy prompt is constructed to match the semantic content expected to emerge at a specific stage of denoising, while ensuring contextual coherence. To construct these proxy prompts, we leverage a large language model (LLM) to analyze the target prompt, identify contradictions, and generate alternative expressions that preserve the original intent while resolving contextual conflicts. By aligning prompt information with the denoising progression, our method enables fine-grained semantic control and accurate image generation in the presence of contextual contradictions. Experiments across a variety of challenging prompts show substantial improvements in alignment to the textual prompt.