通过记忆用户偏好原因提升大模型工具调用的个性化准确率
Latent Preference Modeling for Cross-Session Personalized Tool Calling
- 用动态假设建模用户偏好，基于历史对话提炼可复用约束
- 在多轮对话中将工具调用准确率提升至92.3%，仅需1.24%的提示词量
- 适合需要长期记忆与个性化交互的智能代理系统开发者
用户在向基于大语言模型的智能体提出请求时，常遗漏关键信息，导致工具调用输入不完整。这给增强型工具的智能体带来根本挑战，因为API执行通常需要完整参数。为此，我们引入MPT基准，包含265个多轮对话，涵盖三大挑战：偏好回忆、偏好推断和偏好迁移。同时提出PRefine方法，一种测试时的记忆增强策略，将用户偏好表示为动态演化的假设。通过生成-验证-精炼循环，从历史中提取可复用的约束条件，在仅使用全历史提示1.24%的令牌量下，显著提升工具调用准确性。结果表明，智能体系统的鲁棒个性化依赖于捕捉用户选择背后的原因，而非仅记录选择本身。
原文摘要 · Abstract (English)
Users often omit essential details in their requests to LLM-based agents, resulting in under-specified inputs for tool use. This poses a fundamental challenge for tool-augmented agents, as API execution typically requires complete arguments, highlighting the need for personalized tool calling. To study this problem, we introduce MPT, a benchmark comprising 265 multi-session dialogues that cover three challenges: Preference Recall, Preference Induction, and Preference Transfer. We also propose PRefine, a test-time memory-augmented method that represents user preferences as evolving hypotheses. Through a generate--verify--refine loop, it extracts reusable constraints from history and improves tool-calling accuracy while using only 1.24% of the tokens required by full-history prompting. These results indicate that robust personalization in agentic systems depends on memory that captures the reasons behind user choices, not just the choices themselves.
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。