Building and Comparing Lemma Embeddings for Latin. Classical Latin versus Thomas Aquinas

This paper presents a new set of lemma embeddings for the Latin language. Embeddings are trained on a manually annotated corpus of texts belonging to the Classical era: different models, architectures and dimensions are tested and evaluated using a novel benchmark for the synonym selection task. In addition, we release vectors pre-trained on the “Opera Maiora” by Thomas Aquinas, thus providing a resource to analyze Latin in a diachronic perspective. The embeddings built upon the two training corpora are compared to each other to support diachronic lexical studies. The words showing the highest usage change between the two corpora are reported and a selection of them is discussed.