Collab-REC: An LLM-based Agentic Framework for Balancing Recommendations in Tourism

2025-08-20

AI Summary

To address pervasive popularity bias and overtourism in travel recommendation, this paper proposes a large language model (LLM)-driven multi-agent framework comprising three specialized agents: Personalization, Popularity, and Sustainability. A rule-based, non-LLM coordinator orchestrates multi-round negotiations among them, integrating complementary strategies and incorporating a repeated-response penalty to explicitly model destination carrying capacity and user constraints. Crucially, sustainability is formalized as an explicit optimization objective—enabling tourist load redistribution and discovery of high-quality long-tail destinations. Evaluated on European city recommendation, the framework significantly improves recommendation diversity (+28.6%) and overall relevance (+19.3%) over single-agent baselines, effectively mitigating popularity bias while achieving a tripartite equilibrium among user satisfaction, platform revenue, and destination sustainability.

Technology Category

• Multiagent Systems: Mechanism Design
• Planning, Routing, and Scheduling: Planning with Language Models
• Search and Optimization: Algorithm Configuration

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Agentic search
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems

Abstract

We propose Collab-REC, a multi-agent framework designed to counteract popularity bias and enhance diversity in tourism recommendations. In our setting, three LLM-based agents -- Personalization, Popularity, and Sustainability generate city suggestions from complementary perspectives. A non-LLM moderator then merges and refines these proposals via multi-round negotiation, ensuring each agent's viewpoint is incorporated while penalizing spurious or repeated responses. Experiments on European city queries show that Collab-REC improves diversity and overall relevance compared to a single-agent baseline, surfacing lesser-visited locales that often remain overlooked. This balanced, context-aware approach addresses over-tourism and better aligns with constraints provided by the user, highlighting the promise of multi-stakeholder collaboration in LLM-driven recommender systems.

Problem

• Counteracting popularity bias in tourism recommendations
• Enhancing diversity of suggested destinations
• Balancing multiple stakeholder perspectives collaboratively

Innovation

• Multi-agent LLM framework balances recommendations
• Non-LLM moderator enables multi-round negotiation merging
• Agents represent complementary perspectives: personalization, popularity, sustainability