Loci Similes: A Benchmark for Extracting Intertextualities in Latin Literature

2026-01-12

AI Summary

This study addresses the challenge of intertextuality in Latin literature, where diverse forms and the absence of standardized evaluation benchmarks have hindered the development of automated detection methods. We present Loci Similes, the first large-scale, expert-annotated benchmark dataset comprising approximately 172,000 text passages and 545 verified intertextual pairs, spanning a spectrum from direct quotations to subtle allusions under morphological variation. Leveraging state-of-the-art large language models, we establish semantic similarity–driven retrieval and classification baselines that transcend the limitations of traditional lexical overlap approaches. This work provides the first standardized evaluation platform for intertextuality in Latin literature and significantly advances computational research in classical philology.

Technology Category

• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Analogy

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Web data generation and simulation

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.

Problem

• intertextuality
• Latin literature
• benchmark
• dataset
• text reuse

Innovation

• intertextuality detection
• Latin literature
• benchmark dataset
• semantic similarity
• large language models