Sentence Embedding Models for Ancient Greek Using Multilingual Knowledge Distillation

TL;DR

This paper presents a multilingual knowledge distillation method to develop high-quality sentence embedding models for Ancient Greek, leveraging small parallel datasets and aligning with English translations to enable semantic tasks.

Contribution

It introduces a novel distillation approach and dataset creation for Ancient Greek sentence embeddings, overcoming data scarcity challenges.

Findings

Models perform well on translation search tasks

Effective in semantic similarity and retrieval tasks

Dataset and models are publicly available

Abstract

Contextual language models have been trained on Classical languages, including Ancient Greek and Latin, for tasks such as lemmatization, morphological tagging, part of speech tagging, authorship attribution, and detection of scribal errors. However, high-quality sentence embedding models for these historical languages are significantly more difficult to achieve due to the lack of training data. In this work, we use a multilingual knowledge distillation approach to train BERT models to produce sentence embeddings for Ancient Greek text. The state-of-the-art sentence embedding approaches for high-resource languages use massive datasets, but our distillation approach allows our Ancient Greek models to inherit the properties of these models while using a relatively small amount of translated sentence data. We build a parallel sentence dataset using a sentence-embedding alignment method to align Ancient Greek documents with English translations, and use this dataset to train our models. We evaluate our models on translation search, semantic similarity, and semantic retrieval tasks and investigate translation bias. We make our training and evaluation datasets freely available at https://github.com/kevinkrahn/ancient-greek-datasets .