Loci Similes: A Benchmark for Extracting Intertextualities in Latin Literature

TL;DR

This paper introduces Loci Similes, a comprehensive benchmark dataset designed to facilitate the detection of intertextualities in Latin literature, enabling improved research through standardized evaluation of language models.

Contribution

The paper presents Loci Similes, the first large-scale benchmark dataset for Latin intertextuality detection, along with baseline results using advanced language models.

Findings

Established baseline performance for intertextuality detection

Curated dataset of 172,000 text segments with expert-verified parallels

Demonstrated effectiveness of language models in Latin intertextuality tasks

Abstract

Tracing connections between historical texts is an important part of intertextual research, enabling scholars to reconstruct the virtual library of a writer and identify the sources influencing their creative process. These intertextual links manifest in diverse forms, ranging from direct verbatim quotations to subtle allusions and paraphrases disguised by morphological variation. Language models offer a promising path forward due to their capability of capturing semantic similarity beyond lexical overlap. However, the development of new methods for this task is held back by the scarcity of standardized benchmarks and easy-to-use datasets. We address this gap by introducing Loci Similes, a benchmark for Latin intertextuality detection comprising of a curated dataset of ~172k text segments containing 545 expert-verified parallels linking Late Antique authors to a corpus of classical authors. Using this data, we establish baselines for retrieval and classification of intertextualities with state-of-the-art LLMs.