Large Language Models in Human Mobility and Recommendation Systems

The field of large language models (LLMs) is moving towards more sophisticated applications in human mobility and recommendation systems. Researchers are exploring the use of LLMs to improve semantic understanding of human mobility data, enhance music recommendation systems, and develop more diverse and context-aware tourism recommendations. A key trend is the integration of LLMs with other techniques, such as multi-agent frameworks and memory mechanisms, to improve performance and flexibility. Notable papers include:

  • MobQA, a benchmark dataset for evaluating LLMs on human mobility data, which highlights the limitations of current models in semantic reasoning and explanation.
  • Collab-REC, a multi-agent framework that uses LLMs to balance recommendations in tourism and improve diversity and relevance.
  • RETAIL, a novel dataset and topic-guided multi-agent framework that supports real-world travel planning with implicit queries and environmental awareness.

Sources

MobQA: A Benchmark Dataset for Semantic Understanding of Human Mobility Data through Question Answering

LLM-Based Intelligent Agents for Music Recommendation: A Comparison with Classical Content-Based Filtering

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

Coarse-to-Fine Grounded Memory for LLM Agent Planning

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

Built with on top of