Archimedes NLP Theme Meeting: Invited talk, Monday 3 February, 17:30-19:00 (Greek time)
Speaker: Alessandro Locaputo
Title: "Deep Learning for Digital Humanities: Restoring Latin and Ancient Greek Inscriptions"
Room (BYOL[*]): Pythagoras, Archimedes Unit (1 Artemidos str., ART1 building, ground floor)
and virtually via Zoom Meetings - https://zoom.us/j/96739374236?pwd=mqjih1fkXH6sWl0Gy0qS3KaaEWjAA8.1
• Meeting ID: 967 3937 4236
• Passcode: P2q9kM
Check for additional ways to participate/dial in at the bottom of the announcement.
Abstract:
Inscriptions, artifacts engraved or painted onto stone, are prone to deterioration over time. It is estimated that only 2% to 3% of all Latin inscriptions have survived to the present day, and even these often contain gaps. This makes extracting
information from them a highly challenging task, often requiring meticulous restoration. One possible approach to assist in this restoration process is the application of deep learning techniques.
A significant challenge in working with Latin or Ancient Greek texts, compared to modern English documents, is their scarcity. Furthermore, inscriptions differ considerably in syntax and structure from literary texts, adding another layer of
complexity to applying natural language processing (NLP). Models already trained on these languages cannot be directly applied without significant adjustments.
We analyzed the current state of the art in restoring Latin and Ancient Greek texts, evaluating the performance of existing models across various conditions and proposing new alternatives that incorporate modern Large Language Models (LLMs).
Also, with the goal of making the reasoning processes of these models more interpretable for human users.
As part of our efforts, we initiated a project to combine visual information (e.g., images of inscriptions) with textual data to enhance restoration accuracy. Additionally, we are conducting a survey on intertextuality—the explicit or implicit
relationship between two documents—exploring datasets and techniques for the computational analysis of this phenomenon.
Short Bio:
Alessandro Locaputo is a PhD student in Artificial Intelligence at the University of Udine under the supervision of Prof. Giuseppe Serra and Prof. Andrea Brunello. He holds a Master’s degree in Algorithms and Automatic Reasoning and a Bachelor’s
degree in Computer Science, both from the University of Udine.
The focus of his research is ‘Deep Learning techniques for the restoration of damaged inscriptions,’ with an emphasis on applying natural language processing (NLP) techniques to inscriptions in Latin and Ancient Greek. His work aims to integrate
state-of-the-art artificial intelligence with classical studies, contributing to the preservation and restoration of cultural heritage.
Alessandro has collaborated with the Italian Chamber of Deputies and several local companies on the application of natural language processing, gaining hands-on experience in applying NLP technologies for practical challenges.
Bring Your Own Laptop (BYOL) [*]
We may not be able to assist in-person attendees with audiovisual equipment especially next week. The room (Pythagoras) will be available as usual, but you may need to use your own laptop and headset.
Stay tuned!
For ways to receive news about the NLP Group and its meetings, as well as to get check the latest information about the meetings of Archimedes NLP Theme and AUEB NLP Group, check http://nlp.cs.aueb.gr/news.html. To subscribe to the mailing list
of AUEB NLP Group, send a message with subject "subscribe nodigest" to This email address is being protected from spambots. You need JavaScript enabled to view it.. If you have an AUEB account and want to receive announcements about AUEB NLP Group Meetings through MS Teams, subscribe to "AUEB NLP Group meetings" group
on MS Teams (code: 01j65ny). Team members can also send text messages (chat) to other team members.
If you are an AI researcher or practitioner, please consider becoming a member of the Hellenic Artificial Intelligence Society (EETN, http://www.eetn.gr/en/).
Dial-in information:
One tap mobile
+302111984488,,96739374236#,,,,*318242# Greece
+302311180599,,96739374236#,,,,*318242# Greece
+3902124128823,,96739374236#,,,,*318242# Italy
+390694806488,,96739374236#,,,,*318242# Italy
Dial by your location
• +30 211 198 4488 Greece
• +30 231 118 0599 Greece
• +39 021 241 28 823 Italy
• +39 069 480 6488 Italy
• +39 020 066 7245 Italy
Meeting ID: 967 3937 4236
Passcode: 318242
Find your local number: https://zoom.us/u/adYBDORsI1
