Text-to-Image (TTI) systems are now everyday infrastructure for journalism, education, advertising, and public communication, and the demographic and cultural stereotypes they inherit from training data (rendering women, people of colour, older adults, and non-Western cultures as under-represented or caricatured) become a population-level harm at deployment scale. Existing mitigations either require costly retraining, infeasible for the closed-source backbones that dominate consumer products, or rely on fixed demographic templates that ignore cultural context. We present KG-FairDiff, a model-agnostic, inference-time framework that formalises fairness-aware prompt refinement as a constrained optimisation problem and operationalises it as a closed-loop pipeline: a knowledge graph of ~1,200 culture- and bias-related triples retrieves structured context, an LLM rewriter proposes refinements, and a validator accepts only prompts that reduce a divergence-based fairness loss while preserving semantic fidelity to the user's original intent. We prove a finite-termination bound for the refinement loop, contribute a mathematically consistent evaluation suite linking Bias-P/Bias-W to divergence from target distributions and ENS to KL divergence, and audit eight widely-deployed backbone generators. KG-FairDiff substantially reduces gender, race, age, and intersectional disparities while preserving prompt semantics, offering a practical, deployment-ready route to more equitable generative AI.

翻译：文生图（Text-to-Image, TTI）系统现已成为新闻、教育、广告及公共传播领域的基础设施，其从训练数据中继承的人口统计与文化刻板印象（如将女性、有色人种、老年人和非西方文化呈现为代表性不足或夸张化形象）在规模化部署时会造成群体层面的危害。现有缓解手段要么需要代价高昂的重新训练（对主导消费产品的闭源主干模型不可行），要么依赖忽视文化背景的固定人口统计模板。我们提出KG-FairDiff——一种模型无关的推理时框架，将公平感知提示词优化形式化为带约束的优化问题，并通过闭环流水线实现：包含约1200个文化与偏见相关三元组的知识图谱检索结构化上下文，大语言模型改写器提出优化方案，验证器仅接受既能降低基于散度的公平性损失又保留用户原始意图语义保真度的提示词。我们证明了优化循环的有限终止边界，建立了将Bias-P/Bias-W与目标分布散度关联、ENS与KL散度关联的数学一致性评估套件，并对8个广泛部署的主干生成器进行审计。KG-FairDiff在保持提示语义的同时显著减少了性别、种族、年龄及交叉维度差异，为实现更公平的生成式AI提供了可投入实际部署的实用路径。