ChainRec: An Agentic Recommender Learning to Route Tool Chains for Diverse and Evolving Interests

2026-02-11

AI Summary

This work addresses the limitations of existing large language model–based recommender systems, which typically rely on fixed inference pipelines and struggle to adapt to the diversity and dynamic evolution of user interests—particularly in cold-start or interest-shift scenarios. The authors propose the first recommendation agent framework that enables dynamic tool-chain routing, featuring a standardized tool library and a learnable planner that adaptively selects, sequences, and terminates reasoning tools based on user context. The planner is trained via supervised fine-tuning and preference optimization, while tool standardization is achieved using expert trajectories. Evaluated on multiple datasets in AgentRecBench, the approach significantly improves average HR@{1,3,5}, with especially strong performance in cold-start and interest-evolution settings. Ablation studies further confirm the critical contributions of both tool standardization and preference-based optimization.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Multiagent Systems: Adversarial Agents
• Machine Learning: Learning Preferences or Rankings

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Large language models (LLMs) are increasingly integrated into recommender systems, motivating recent interest in agentic and reasoning-based recommendation. However, most existing approaches still rely on fixed workflows, applying the same reasoning procedure across diverse recommendation scenarios. In practice, user contexts vary substantially-for example, in cold-start settings or during interest shifts, so an agent should adaptively decide what evidence to gather next rather than following a scripted process. To address this, we propose ChainRec, an agentic recommender that uses a planner to dynamically select reasoning tools. ChainRec builds a standardized Tool Agent Library from expert trajectories. It then trains a planner using supervised fine-tuning and preference optimization to dynamically select tools, decide their order, and determine when to stop. Experiments on AgentRecBench across Amazon, Yelp, and Goodreads show that ChainRec consistently improves Avg HR@{1,3,5} over strong baselines, with especially notable gains in cold-start and evolving-interest scenarios. Ablation studies further validate the importance of tool standardization and preference-optimized planning.

Problem

• recommender systems
• large language models
• agentic reasoning
• cold-start
• evolving interests

Innovation

• agentic recommendation
• dynamic tool routing
• tool standardization
• preference optimization
• reasoning-based recommender