Over the last two years, the spread of Generative AI has opened new opportunities in all the fields. The overall impression is that it could speed up different operations related to content generation, such as narratives and data-driven stories. The article focuses on applying Generative AI to the specific domain of Archaeology. Starting from raw field data and specialized publications, this paper aims to verify the impact of scientific data in the construction of fictional data-driven stories. The proposed system, based on Retrieval Augmented Generation (RAG), combines the archaeological documents and two popular Large Language Models (LLMs): GPT-3.5-turbo and GPT-4.0. Preliminary results show that the implemented system can build quite satisfactory narratives.
Combining Generative AI and Archaeology to Build Data-Driven Stories
Francesca Buscemi
;Angelica Lo Duca
2024
Abstract
Over the last two years, the spread of Generative AI has opened new opportunities in all the fields. The overall impression is that it could speed up different operations related to content generation, such as narratives and data-driven stories. The article focuses on applying Generative AI to the specific domain of Archaeology. Starting from raw field data and specialized publications, this paper aims to verify the impact of scientific data in the construction of fictional data-driven stories. The proposed system, based on Retrieval Augmented Generation (RAG), combines the archaeological documents and two popular Large Language Models (LLMs): GPT-3.5-turbo and GPT-4.0. Preliminary results show that the implemented system can build quite satisfactory narratives.File | Dimensione | Formato | |
---|---|---|---|
AIUCD 2024 Buscemi-Lo Duca.pdf
accesso aperto
Licenza:
Creative commons
Dimensione
5.56 MB
Formato
Adobe PDF
|
5.56 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.