How much do Large Language Models know about panel paintings preservation?
How much do Large Language Models know about panel paintings preservation?
This study evaluates the capability of large language models (LLMs) in understanding the preservation of paintings on panel by comparison with the predictions obtained through the digital platform HERIe. The latter is specialized tool for heritage object risk assessment. Four large language models (LLMs) - ChatGPT 3.5, ChatGPT 4, Claude, and Gemini - were tested asking what are the levels of strain experienced by panel paintings under different conditions. The models were also tested on their ability to rank different environments conditions in order of suitability for storing panel paintings and were examined whether the languages of prompts affected results. The study concludes that while LLMs demonstrate a general understanding of wood panel preservation principles, they lack the specialized calculation abilities of purpose-built tools like HERIe for precise risk assessment in cultural heritage preservation.
Li, Fengjun
bcf63cba-f1fd-48fe-bc5a-219e548d7a42
Reggio, Daniela
bb19bbfe-dad9-4f94-97d8-6d78f2678231
Grau-Bove, Josep
21a889cd-064b-4e48-8997-61b7a3f29784
Gauthier, Stephanie
4e7702f7-e1a9-4732-8430-fabbed0f56ed
Nicol, Fergus
55e3b6e4-885d-4aa4-96a8-441ed11e1eaa
Brotas, Luisa
44ab859c-b1ab-40a3-aedf-82d4f7624f09
Altamirano, Hector
9c06526d-78ab-451f-9dcd-0211a3d220ed
16 September 2024
Li, Fengjun
bcf63cba-f1fd-48fe-bc5a-219e548d7a42
Reggio, Daniela
bb19bbfe-dad9-4f94-97d8-6d78f2678231
Grau-Bove, Josep
21a889cd-064b-4e48-8997-61b7a3f29784
Gauthier, Stephanie
4e7702f7-e1a9-4732-8430-fabbed0f56ed
Nicol, Fergus
55e3b6e4-885d-4aa4-96a8-441ed11e1eaa
Brotas, Luisa
44ab859c-b1ab-40a3-aedf-82d4f7624f09
Altamirano, Hector
9c06526d-78ab-451f-9dcd-0211a3d220ed
Li, Fengjun, Reggio, Daniela and Grau-Bove, Josep
(2024)
How much do Large Language Models know about panel paintings preservation?
Gauthier, Stephanie, Nicol, Fergus, Brotas, Luisa and Altamirano, Hector
(eds.)
13th Masters Conference: People and Buildings, London, London, United Kingdom.
16 Sep 2024.
6 pp
.
(doi:10.5258/SOTON/P1207).
Record type:
Conference or Workshop Item
(Paper)
Abstract
This study evaluates the capability of large language models (LLMs) in understanding the preservation of paintings on panel by comparison with the predictions obtained through the digital platform HERIe. The latter is specialized tool for heritage object risk assessment. Four large language models (LLMs) - ChatGPT 3.5, ChatGPT 4, Claude, and Gemini - were tested asking what are the levels of strain experienced by panel paintings under different conditions. The models were also tested on their ability to rank different environments conditions in order of suitability for storing panel paintings and were examined whether the languages of prompts affected results. The study concludes that while LLMs demonstrate a general understanding of wood panel preservation principles, they lack the specialized calculation abilities of purpose-built tools like HERIe for precise risk assessment in cultural heritage preservation.
Text
MC2024KT8011_Fengjun Li
- Version of Record
More information
Published date: 16 September 2024
Venue - Dates:
13th Masters Conference: People and Buildings, London, London, United Kingdom, 2024-09-16 - 2024-09-16
Identifiers
Local EPrints ID: 494898
URI: http://eprints.soton.ac.uk/id/eprint/494898
PURE UUID: c7074d6c-e611-4c77-833f-17d1d65e6b43
Catalogue record
Date deposited: 22 Oct 2024 16:36
Last modified: 23 Oct 2024 01:47
Export record
Altmetrics
Contributors
Author:
Fengjun Li
Author:
Daniela Reggio
Author:
Josep Grau-Bove
Editor:
Fergus Nicol
Editor:
Luisa Brotas
Editor:
Hector Altamirano
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
