AI 中文总结
CraftAlign是对齐AI故事与人类叙事技巧的框架，含特征估计器与能量模型，可评估写作模式并生成修订指导，实验显示其区分能力及指导效果优于基线。
AI 中文摘要
大型语言模型如今可生成流畅完整的故事，但许多输出仍因陈词滥调、过度解释、线性因果推进及刻板结局而显得程式化、不自然，带有可立即识别的AI痕迹。现有检测与评估方法常止步于来源标签或整体评分，而修订方法通常针对预定义问题进行局部编辑，限制了其支持多种合理修订策略或指导故事层面的信息发布、因果组织及结局处理变化的能力。我们提出CraftAlign，这一框架通过评估人类/AI写作模式并提供修订指导，使AI故事与人类讲故事的技巧对齐。CraftAlign包含两个学习模块和一个推理时指导流程：基于Qwen3.5-9B构建的特征估计器预测涵盖风格与叙事的304项显式写作特征；类别条件能量模型根据原始写作提示（若可用），对所得特征配置与人类及AI写作模式进行评分。推理时，CraftAlign应用符合 schema 的结构化扰动，选择使特征配置向人类写作模式靠拢的变化，并将其转换为自然语言指导，供独立编辑重写完整故事。实验表明，CraftAlign可准确区分人类与AI写作模式，且其指导在编辑及人类研究中均优于修订基线。
英文摘要
Large language models can now generate fluent and complete stories, yet many outputs still feel formulaic and unnatural because of cliches, over-explanation, linear causal progression, and stereotyped endings, an immediately recognizable AI flavor. Existing detection and evaluation methods often stop at source labels or holistic scores, while revision methods typically target predefined issues through localized edits, limiting their ability to support multiple plausible revision strategies or guide story-wide changes in information release, causal organization, and ending treatment. We introduce CraftAlign, a framework that aligns AI stories with the craft of human storytelling by both assessing Human/AI writing patterns and providing revision guidance. CraftAlign comprises two learned modules and an inference-time guidance pipeline. A feature estimator built on Qwen3.5-9B predicts 304 explicit writing features spanning style and narrative. A class-conditional energy model scores the resulting feature configuration against Human and AI writing patterns, conditioning on the original writing prompt when available. At inference time, CraftAlign applies schema-valid structured perturbations, selects changes that move the feature configuration toward the Human writing pattern, and converts them into natural-language guidance for a separate editor to rewrite the full story. Experiments show that CraftAlign accurately distinguishes Human and AI writing patterns and that its guidance outperforms revision baselines across editors and in a human study.
Comments17 pages, 5 figures, includes appendix
