TourPlanner: A Competitive Consensus Framework with Constraint-Gated Reinforcement Learning for Travel Planning

2026-01-08

arXiv.org

AI Summary

This work proposes TourPlanner, a novel framework addressing key challenges in automated tour planning: the imbalance between candidate point-of-interest (POI) recall and filtering efficiency, limited exploration of the solution space due to single-path reasoning, and the difficulty of jointly optimizing hard and soft constraints. TourPlanner integrates a Personalized Recall and Spatial Optimization (PReSO) pipeline, a Competitive Consensus Chain-of-Thought (CCoT) multi-path reasoning mechanism, and a Sigmoid-based constraint-gated reinforcement learning strategy. This design enables strict adherence to hard constraints while dynamically optimizing user preferences through soft constraints. Experimental results on standard tour planning benchmarks demonstrate that TourPlanner significantly outperforms existing methods, achieving state-of-the-art performance in both itinerary feasibility and alignment with user preferences.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Travel planning is a sophisticated decision-making process that requires synthesizing multifaceted information to construct itineraries. However, existing travel planning approaches face several challenges: (1) Pruning candidate points of interest (POIs) while maintaining a high recall rate; (2) A single reasoning path restricts the exploration capability within the feasible solution space for travel planning; (3) Simultaneously optimizing hard constraints and soft constraints remains a significant difficulty. To address these challenges, we propose TourPlanner, a comprehensive framework featuring multi-path reasoning and constraint-gated reinforcement learning. Specifically, we first introduce a Personalized Recall and Spatial Optimization (PReSO) workflow to construct spatially-aware candidate POIs'set. Subsequently, we propose Competitive consensus Chain-of-Thought (CCoT), a multi-path reasoning paradigm that improves the ability of exploring the feasible solution space. To further refine the plan, we integrate a sigmoid-based gating mechanism into the reinforcement learning stage, which dynamically prioritizes soft-constraint satisfaction only after hard constraints are met. Experimental results on travel planning benchmarks demonstrate that TourPlanner achieves state-of-the-art performance, significantly surpassing existing methods in both feasibility and user-preference alignment.

Problem

• travel planning
• points of interest
• constraint optimization
• multi-path reasoning
• feasible solution space

Innovation

• multi-path reasoning
• constraint-gated reinforcement learning
• competitive consensus
• travel planning
• hard and soft constraints