LLMs as Orchestrators: Constraint-Compliant Multi-Agent Optimization for Recommendation Systems

Recommendation systems must optimize multiple objectives while satisfying hard business constraints such as fairness and coverage. For example, an e-commerce platform may require every recommendation list to include items from multiple sellers and at least one newly listed product; violating such constraints–even once–is unacceptable in production. Prior work on multi-objective recommendation and recent LLM-based recommender agents largely treat constraints as soft penalties or focus on item scoring and interaction, leading to frequent violations in real-world deployments. How to leverage LLMs for coordinating constrained optimization in recommendation systems remains underexplored. We propose DualAgent-Rec, an LLM-coordinated dual-agent framework for constrained multi-objective e-commerce recommendation. The framework separates optimization into an Exploitation Agent that prioritizes accuracy under hard constraints and an Exploration Agent that promotes diversity through unconstrained Pareto search. An LLM-based coordinator adaptively allocates resources between agents based on optimization progress and constraint satisfaction, while an adaptive epsilon-relaxation mechanism guarantees feasibility of final solutions. Experiments on the Amazon Reviews 2023 dataset demonstrate that DualAgent-Rec achieves 100% constraint satisfaction and improves Pareto hypervolume by 4-6% over strong baselines, while maintaining competitive accuracy-diversity trade-offs. These results indicate that LLMs can act as effective orchestration agents for deployable and constraint-compliant recommendation systems.

💡 Research Summary

The paper tackles a pressing problem in e‑commerce recommendation: simultaneously optimizing multiple objectives—relevance, diversity, novelty—while strictly satisfying hard business constraints such as seller coverage, category fairness, and minimum new‑item exposure. Existing multi‑objective recommendation methods either treat these constraints as soft penalties or rely on post‑hoc filtering, which leads to frequent violations in production. To bridge this gap, the authors introduce DualAgent‑Rec, a novel framework that leverages a large language model (LLM) as a high‑level coordinator for constrained multi‑objective optimization.

Problem formulation: Given a user’s interaction history, an item catalog, and item feature vectors, the task is to select a fixed‑size recommendation list L that maximizes a three‑dimensional objective vector f(L) =