arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.09829cs.IR
收藏

内部化多智能体推理以实现准确且高效的基于LLM的推荐

Internalizing Multi-Agent Reasoning for Accurate and Efficient LLM-based Recommendation

Tencent(腾讯)

Yang Wu, Haoze Wang, Qian Li, Jun Zhang, Huan Yu, Jie Jiang

AI总结:

本文提出STAR模型,通过轨迹驱动的内部化方法,将多智能体推理能力高效整合到单模型中,实现准确且高效的推荐系统。

摘要:

大型语言模型(LLM)通过利用广泛的世界知识和语义推理来解读用户意图,正在重塑推荐系统。然而,有效整合这些能力与协作信号的同时避免不利的推理延迟仍然是一个关键瓶颈。为此,我们提出了一种轨迹驱动的内部化框架,以开发一个单智能体轨迹对齐推荐器(STAR)。具体来说,为了将复杂的推理能力内部化到一个高效模型中,我们首先设计了一个多智能体教师系统,能够进行多轮工具使用和反思。该教师利用协作信号翻译机制,将潜在的行为模式显式地转换为描述性自然语言证据,以提高推理准确性。随后,轨迹驱动的蒸馏管道将这种智能逻辑,包括规划、工具使用和自我反思,转移到紧凑的STAR模型中。广泛的实验表明,STAR在消除了迭代延迟的同时,超越其教师系统8.7%至39.5%,为实时、基于推理的推荐铺平了道路。

英文摘要:

Large Language Models (LLMs) are reshaping recommender systems by leveraging extensive world knowledge and semantic reasoning to interpret user intent. However, effectively integrating these capabilities with collaborative signals while avoiding prohibitive inference latency remains a critical bottleneck. To address this, we propose a trajectory-driven internalization framework to develop a Single-agent Trajectory-Aligned Recommender (STAR). Specifically, to internalize complex reasoning capabilities into a single efficient model, we first design a multi-agent teacher system capable of multi-turn tool usage and reflection. This teacher utilizes a Collaborative Signal Translation mechanism to explicitly convert latent behavioral patterns into descriptive natural language evidence to enhance reasoning accuracy. Subsequently, a trajectory-driven distillation pipeline transfers this agentic logic, including planning, tool usage, and self-reflection, into the compact STAR model. Extensive experiments demonstrate that STAR surpasses its teacher by 8.7% to 39.5% while eliminating iterative latency, paving the way for real-time, reasoning-enhanced recommendation.

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。

↑