ACL Findings2026
Internalizing Multi-Agent Reasoning for Accurate and Efficient LLM-based Recommendation
Yang Wu, Haoze Wang, Qian Li, Jun Zhang, Huan Yu, Jie Jiang
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.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- LLM-ESR: Large Language Models Enhancement for Long-tailed Sequential Recommendation cites Representation Learning with Large Language Models for Recommendation
- Representation Learning with Large Language Models for Recommendation cites LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
Builds on10
- LightGCN: Simplifying and Powering Graph Convolution Network for RecommendationXiangnan He, Kuan Deng, Xiang Wang, Yan Li et al.SIGIR 2020 · 4,448 citations
- Representation Learning with Large Language Models for RecommendationXubin Ren, Wei Wei, Lianghao Xia, Lixin Su et al.WWW 2024 · 385 citations
- LLM-ESR: Large Language Models Enhancement for Long-tailed Sequential RecommendationQidong Liu, Xian Wu, Yejing Wang, Zijian Zhang et al.NeurIPS 2024 · 154 citations
- Collaborative Large Language Model for Recommender SystemsYaochen Zhu, Liang Wu, Qi Guo, Liangjie Hong et al.WWW 2024 · 150 citations
- HybridFlow: A Flexible and Efficient RLHF FrameworkGuangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu et al.EuroSys 2025 · 61 citations
Related papers
- Think, But Don't Tell: Implicit Reasoning for LLM-based Sequential Recommendation via Multi-Teacher DistillationWeihai Lu, Xiaoxi Cui, Chenke YinSIGIR 2026
- Tool-Star: Empowering Multi-Tool Collaborative Web Agent via Reinforcement LearningGuanting Dong, Yifei Chen, Xiaoxi Li, Jiajie Jin et al.SIGIR 2026 · 1 citation
- Distilling LLM Reasoning into Dense Encoders: Bridging the Accuracy-Efficiency Gap in RecommendationDonghee Han, Daeyoung Roh, A Young Kim, Hwanjun Song et al.ACL Findings 2026
- Token-Efficient Long-Term Interest Sketching and Internalized Reasoning for LLM-based RecommendationZhihao Ding, Jinming Li, Shuai Mu, Jieming ShiICLR 2026
- SmartAD: Capacity-Aligned Agent Distillation for Small Language ModelsGuokai Tang, Feng ZhaoACL Findings 2026
