解决个性化模型中风格指令与用户偏好冲突问题
Do Implicit Personalization and Explicit Styles Conflict? PsPLUG: A Lightweight Plug-in for Balancing Personalization and Style in Customized LLMs
- 通过残差学习在风格控制后保留用户特定偏好
- 实验显示传统方法风格指令会削弱个性化，而新方法可平衡两者
- 轻量级插件设计，推理时可调个性化强度，适合定制化场景
个性化大语言模型通常需要遵循明确的风格指令，但我们发现这些指令可能损害个性化方法本应保持的用户特定特征。我们称此现象为个性化崩溃：显式风格控制与隐式用户偏好存在冲突。为此，我们提出PsPLUG，一种轻量级插件，它在考虑请求风格后学习用户特定的残差。PsPLUG还支持在推理时调节个性化强度。实验表明，显式风格指令会削弱现有方法的个性化表现，而PsPLUG能更好保留用户偏好，同时实现对个性化与风格遵循之间平衡的精确控制。
原文摘要 · Abstract (English)
Personalized large language models are often expected to follow explicit style instructions, yet we find that such instructions can undermine the user-specific characteristics that personalization methods aim to preserve. We call this failure mode personalization collapse: explicit style control can conflict with implicit user preferences. To address this challenge, we propose PsPLUG, a lightweight plug-in that learns a user-specific residual after accounting for the requested style. PsPLUG also allows us to tune personalization strength at inference time. Our experiments show that explicit style instructions can diminish personalization in existing methods, whereas PsPLUG better preserves user preferences while providing precise control over the balance between personalization and style adherence.
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。