arXivDaily arXiv每日学术速递 周一至周五更新
​通知：联邦节假日调休：哥伦布日 / 原住民日（Columbus Day / Indigenous Peoples' Day），今日（13日）无更新。
arXiv 2503.02450cs.CL

衡量让你独特之处：用于增强 LLM 个性化的差异感知用户建模

Measuring What Makes You Unique: Difference-Aware User Modeling for Enhancing LLM Personalization

National University of Singapore（新加坡国立大学）；The Chinese University of Hong Kong（香港中文大学）；University of Science and Technology of China（中国科学技术大学）

Yilun Qiu, Xiaoyan Zhao, Yang Zhang, Yimeng Bai, Wenjie Wang, Hong Cheng, Fuli Feng, Tat-Seng Chua

备注：
2025 ACL Findings
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对现有 LLM 个性化忽视用户间差异的问题，提出 DPL，通过代表性用户比较与结构化差异提取生成偏好上下文，并在真实数据集上验证其显著提升效果。

摘要：

个性化大型语言模型（LLM）已成为推动其广泛应用以改善个人生活体验的关键步骤。在追求个性化的过程中，从个人历史数据中提炼关键偏好信息，并将其作为指令式偏好上下文以定制 LLM 生成，已成为一个有前景的方向。然而，这些方法存在一个根本性局限：忽视了用户间比较分析，而这种分析对于识别真正塑造偏好的用户间差异至关重要。为解决该局限，我们提出 Difference-aware Personalization Learning（DPL，差异感知个性化学习），这是一种强调提取用户间差异以增强 LLM 个性化的新方法。DPL 会策略性地选择代表性用户进行比较，并建立结构化标准，以提取有意义且与任务相关的差异，用于定制 LLM 生成。在真实世界数据集上的大量实验表明，DPL 显著增强了 LLM 个性化。我们在 https://github.com/SnowCharmQ/DPL 发布代码。

英文摘要：

Personalizing Large Language Models (LLMs) has become a critical step in facilitating their widespread application to enhance individual life experiences. In pursuit of personalization, distilling key preference information from an individual's historical data as instructional preference context to customize LLM generation has emerged as a promising direction. However, these methods face a fundamental limitation by overlooking the inter-user comparative analysis, which is essential for identifying the inter-user differences that truly shape preferences. To address this limitation, we propose Difference-aware Personalization Learning (DPL), a novel approach that emphasizes extracting inter-user differences to enhance LLM personalization. DPL strategically selects representative users for comparison and establishes a structured standard to extract meaningful, task-relevant differences for customizing LLM generation. Extensive experiments on real-world datasets demonstrate that DPL significantly enhances LLM personalization. We release our code at https://github.com/SnowCharmQ/DPL.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑