Year: 2026
Type: conference-paper
Abstract: 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 trajectorydriven 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 trajectorydriven 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.
Authors: Yang Wu, Haoze Wang, Qian Li, Jun Zhang, Huan Yu, Jie Jiang
Institution: Tencent (China)
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166870181