用高效微调让大模型模仿十位作者的写作风格
Customizing Large Language Model Generation Style using Parameter-Efficient Finetuning
- 用低秩适配微调技术，仅少量参数即可改变大模型风格
- 生成文本在词汇、句法和表面特征上与目标作者高度一致
- 适合需要个性化写作助手的用户，但难以记住具体内容
通用大语言模型（LLM）正被广泛用于辅助写作，但其固有的写作风格未必适合所有用户或场景。若能将模型的个人化写作风格进行定制，将极大提升其作为写作助手的价值。本文探索了参数高效微调（PEFT）中的低秩适配（LoRA）方法，是否能有效引导大模型生成特定风格的内容。我们使用该方法对LLaMA-2模型进行了微调，使其模仿十位不同作者的写作风格。结果表明，生成文本在词汇选择、句法结构和表面特征上均与目标作者高度一致，但在内容记忆能力方面表现较弱。研究验证了PEFT在实现用户级风格定制方面的可行性与潜力。
原文摘要 · Abstract (English)
One-size-fits-all large language models (LLMs) are increasingly being used to help people with their writing. However, the style these models are trained to write in may not suit all users or use cases. LLMs would be more useful as writing assistants if their idiolect could be customized to match each user. In this paper, we explore whether parameter-efficient finetuning (PEFT) with Low-Rank Adaptation can effectively guide the style of LLM generations. We use this method to customize LLaMA-2 to ten different authors and show that the generated text has lexical, syntactic, and surface alignment with the target author but struggles with content memorization. Our findings highlight the potential of PEFT to support efficient, user-level customization of LLMs.
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。