Logion: Machine Learning for Greek Philology

Charlie Cowen-Breen, Creston Brooks, Johannes Haubold, Barbara Graziosi
5/1/2023
cs.CLcs.LG

No official code found. Best implementation: princeton-logion/logion-app (5 stars, MIT, last commit 2025-12-24).

Abstract

This paper presents machine-learning methods to address various problems in Greek philology. After training a BERT model on the largest premodern Greek dataset used for this purpose to date, we identify and correct previously undetected errors made by scribes in the process of textual transmission, in what is, to our knowledge, the first successful identification of such errors via machine learning. Additionally, we demonstrate the model's capacity to fill gaps caused by material deterioration of premodern manuscripts and compare the model's performance to that of a domain expert. We find that best performance is achieved when the domain expert is provided with model suggestions for inspiration. With such human-computer collaborations in mind, we explore the model's interpretability and find that certain attention heads appear to encode select grammatical features of premodern Greek.

View on arXivView PDF

Code Implementations(1)

princeton-logion/logion-appCommunity72%
MIT

Machine learning for Greek philology

50CSS, HTMLJan 2, 202510 months agoMIT

Similar Papers

An Annotated Dataset of Errors in Premodern Greek and Baselines for Detecting Them

85% match1 repo32024

Instruct-Tuning Pretrained Causal Language Models for Ancient Greek Papyrology and Epigraphy

81% match2024

The Patrologia Graeca Corpus: OCR, Annotation, and Open Release of Noisy Nineteenth-Century Polytonic Greek Editions

81% match2026

Contrastive Learning for Character Detection in Ancient Greek Papyri

80% match2024

NLP for The Greek Language: A Longer Survey

79% match2024

Forging GEMs: Advancing Greek NLP through Quality-Based Corpus Curation

79% match2025

Cite this paper

@article{cowenbreen2023logion,
  title  = {Logion: Machine Learning for Greek Philology},
  author = {Charlie Cowen-Breen and Creston Brooks and Johannes Haubold and Barbara Graziosi},
  year   = {2023},
  eprint = {2305.01099},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2305.01099v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2305.01099.svg)](https://codeofpaper.com/paper/2305.01099)

Discussion