LLM-Powered Context-Aware Guide Enhances Triangolazioni Cultural Heritage Webapp

A new research paper introduces an extension of Triangolazioni, a Cultural Heritage webapp, that integrates Large Language Models (LLMs) to provide context-dependent, external information. The system is built on a loosely-coupled architecture that is agnostic to the specific LLM used, allowing for flexible integration. The extension enriches curated content with dynamic, context-aware information, supporting both search and presentation of relevant data. The paper, titled 'A Context-Aware Cultural Heritage Guide Powered by LLMs,' was submitted to the Computer Science > Information Retrieval section on arXiv (arXiv:2608.00549). The work highlights the potential of LLMs to enhance cultural heritage experiences by offering personalized and contextually relevant information. The submission history and additional features are available on the arXiv platform, including references, citations, and tools for bibliographic management. The project aligns with arXivLabs, a framework for experimental projects, emphasizing values of openness, community, excellence, and user data privacy. The research demonstrates a practical application of AI in the cultural sector, potentially transforming how users engage with heritage content.