
Google DeepMind’s Aeneas Model Revolutionizes Historical Text Restoration
News Summary
Google DeepMind has developed an AI model named Aeneas, designed to restore and contextualize ancient Latin inscriptions. The model, which is open-source and free to use, assists historians by identifying parallels across thousands of Latin texts, thus aiding in the reconstruction of fragmented inscriptions. Aeneas is capable of recognizing deep connections and offering interpretable suggestions, serving as a vital tool within historians’ workflows. The model can restore textual gaps of unknown length and operates using both textual and visual inputs. Additionally, Aeneas has enhanced the capabilities of Ithaca, another DeepMind model focused on Ancient Greek texts. Researchers can explore the model on DeepMind’s “Predicting the Past” website.




