RETAIL: Towards Real-world Travel Planning for Large Language Models

2025-08-21

AI Summary

Current large language models face three key bottlenecks in real-world travel planning: difficulty in identifying implicit user needs, insufficient modeling of environmental constraints and personal preferences, and coarse-grained, detail-deficient itinerary generation. To address these challenges, we propose TGMA—the first travel planning framework supporting implicit query understanding and environment-aware reasoning—built upon a topic-guided multi-agent collaboration mechanism. TGMA jointly integrates implicit intent recognition, dynamic environment modeling, personalized preference inference, and fine-grained point-of-interest (POI) integration to enable end-to-end, executable planning. We further introduce RETAIL, a comprehensive benchmark dataset encompassing diverse query types and interactive revision scenarios. Experiments on real-world tasks show that state-of-the-art models achieve only a 1.0% success rate, while TGMA improves this to 2.72%, significantly enhancing plan feasibility and practical utility—establishing a new paradigm for deployable travel planning systems.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Multiagent Systems: Multiagent Planning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

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

Abstract

Although large language models have enhanced automated travel planning abilities, current systems remain misaligned with real-world scenarios. First, they assume users provide explicit queries, while in reality requirements are often implicit. Second, existing solutions ignore diverse environmental factors and user preferences, limiting the feasibility of plans. Third, systems can only generate plans with basic POI arrangements, failing to provide all-in-one plans with rich details. To mitigate these challenges, we construct a novel dataset extbf{RETAIL}, which supports decision-making for implicit queries while covering explicit queries, both with and without revision needs. It also enables environmental awareness to ensure plan feasibility under real-world scenarios, while incorporating detailed POI information for all-in-one travel plans. Furthermore, we propose a topic-guided multi-agent framework, termed TGMA. Our experiments reveal that even the strongest existing model achieves merely a 1.0% pass rate, indicating real-world travel planning remains extremely challenging. In contrast, TGMA demonstrates substantially improved performance 2.72%, offering promising directions for real-world travel planning.

Problem

• Addressing implicit user queries in travel planning systems
• Incorporating environmental factors for realistic plan feasibility
• Generating detailed all-in-one travel plans beyond basic POIs

Innovation

• Topic-guided multi-agent framework
• Dataset for implicit query decision-making
• Environmental awareness for plan feasibility