6.04 RT – AI and Byzantine Studies

Computer-assisted Byzantine Studies still mainly equal the creation of simple databases which, in the best of cases, provide some openings for the linking of data. Only little is done to take recent developments within AI into account (for instance, to engage with neural networks, deep learning, machine reading, and generative AI), and to discuss what impact all of this is bound to have on the discipline in coming years. Indeed, there is reason to believe that Byzantines Studies is lagging behind comparable disciplines in this development. The aim of this RT is to gather scholars from within and without Byzantine Studies to discuss practices and prospects. The speakers will make short presentations focusing on the potential for applying AI within their own specialities: the study of text-bearing objects (such as seals, inscriptions, and manuscripts), cultural records (such as history, literature, archeology and impacted space), and linguistics. After the presentations, there will be a general discussion.

Philology Meets Machine Learning: Digital Annotation and Semantic Modelling of Byzantine Greek Texts

Diplomatic transcriptions of medieval texts pose particular challenges for digital annotation: their fragmentary preservation, non-standard orthography, and heterogeneous linguistic features complicate processing and limit the reuse of existing tools. To address these issues, we developed a full annotation pipeline for unedited medieval Greek, including part-of-speech tagging, morphological analysis, and lemmatisation. A manually created gold standard…

Read More Philology Meets Machine Learning: Digital Annotation and Semantic Modelling of Byzantine Greek Texts
AI and Byzantine Studies: Structures of Knowledge at Scale

The central methodological goal of the ERC project RELEVEN (2021–26) has been to represent data about the history of the medieval Near East, including Byzantine history, in a way that meets the needs of the scholars who study that history – this means encoding the conflicting accounts of events, room for doubt concerning facts, and…

Read More AI and Byzantine Studies: Structures of Knowledge at Scale
Byzantine Sigillographic Knowledge as Linked Open Data: Dos and Don’ts

This paper discusses the potential and the challenges involved in modelling Byzantine sigillographic knowledge as Linked Open Data. The fast-growing availability of digital resources on seals has created new opportunities for the integration of prosopographical, institutional and archaeological information. Recent advances in artificial intelligence further intensify these possibilities by enabling automated entity recognition, pattern detection…

Read More Byzantine Sigillographic Knowledge as Linked Open Data: Dos and Don’ts
Machine Learning for the Humanities

Recent advances in machine learning led to the development of large-scale models, trained on diverse corpora. These models hold immense potential for analyzing historical texts, cultural artifacts, and long-term societal trends, yet their applicability to humanistic inquiry is limited by three key challenges: the temporal drift of language across centuries, the subjectivity of interpretation when…

Read More Machine Learning for the Humanities
Who Murdered Emperor Nikephoros I? An Experiment in Generating Data on Political Instability in the Byzantine Empire with the Help of Large Language Models (LLMs)

One of the most widely used and best-known type of “Artificial Intelligence” are Large Language Models (LLMs), trained on massive amounts of text data to process and generate human language (such as ChatGPT or Google Gemini). In an experiment executed by the SESHAT Global History Databank at the Complexity Science Hub in Vienna in cooperation…

Read More Who Murdered Emperor Nikephoros I? An Experiment in Generating Data on Political Instability in the Byzantine Empire with the Help of Large Language Models (LLMs)