用小模型+LoRA让AI改写文本更像用户本人，无需提示词。
Small Is Enough: Per-User Style Rewriting of AI-Edited Text via LoRA Adapters
- 用多个小模型生成配对数据，微调LoRA适配器匹配用户风格。
- 在科学论文上测试，小模型表现已达上限，0.69分（满分1）。
- 用户风格更自然，大模型虽好但小模型已够用，适合隐私敏感场景。
InMyStyle 是一个以隐私为先的单用户系统，无需推理时的指令提示，即可将小型语言模型适配至重写由AI修改过的文本，使其更贴近特定用户的写作风格。基于用户文档，系统使用多个本地辅助LLM构建成对训练样本，并在0.5B至7B参数的基模型上微调LoRA适配器。通过长度感知生成预算与自动分块机制，支持不同长度输入。在来自科学论文语料库的219组评估对上，无论采用贪婪解码还是采样解码，自动综合得分均稳定在0.69（0-1量表）。该得分平台现象表明：小模型已足够完成此重写任务，模型规模仅影响权衡而非质量排名。作为次要评估，5名LLM评审员对400条输出进行评分，InMyStyle生成内容的平均感知AI度比其辅助生成输入降低超20%，且在InMyStyle内部，模型越大，感知AI度越低。
原文摘要 · Abstract (English)
InMyStyle is a privacy first, single user system that adapts small language models to rewrite AI-edited text towards an individual user's writing style without an instruction prompt at inference. Given a user's documents, it uses multiple local helper LLMs to construct paired training examples and fine tunes LoRA adapters on base models ranging from 0.5B to 7B parameters. Length aware generation budgets and automatic chunking support inputs of different lengths. On 219 evaluation pairs from a scientific-paper corpus, the automatic composite score plateaus at 0.69 [scale 0-1] across all model sizes under both greedy and sampled decoding. This observed plateau suggests that small models are sufficient for the measured rewriting task, with model size determining trade-offs rather than a stable quality ranking. As a secondary evaluation, 400 ratings from five LLM judges give InMyStyle outputs a mean perceived AI-ness score over 20% lower than their helper-AI generated inputs, while mean perceived AI-ness scores decrease with model size within InMyStyle.
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。