We present the first contextualization model for antique inscriptions, intended to aid historians better interpret, attribute, and recover fragmentary texts.
Writing was everywhere in the Roman world – carved on everything from imperial monuments to everyday objects. From political graffiti, love poems and epitaphs to business transactions, birthday invitations and magical spells, the inscriptions offer state-of-the-art historians a luxurious insight into the diversity of everyday life in the Roman world.
Often these texts are fragmentary, faded or deliberately destroyed. They are almost impossible to restore, date and place without contextual information, especially when comparing similar inscriptions.
Today we publish approx paper in Nature introduction Aeneasthe first artificial intelligence (AI) model to contextualize antique inscriptions.
When working with antique inscriptions, historians traditionally rely on their knowledge and specialized resources to identify “similarities” – that is, texts that are similar in sound, syntax, standardized formulas, and origin.
Aeneas significantly speeds up this elaborate and time-consuming work. It analyzes thousands of Latin inscriptions, finding textual and contextual similarities in seconds, allowing historians to interpret and build on the model’s findings.