Neural Machine Translation (NMT) and Large Language Models (LLM) Applied to Scientific Research in Comparative Literature: Which Models for Which Practices?

Abstract

Research in French as a Foreign Language (FLE) is evolving thanks to advances in artificial intelligence, particularly with neural machine translation (NMT) and large language models, which ease access to foreign languages and cultures. In Algeria, research in comparative literature remains constrained and requires profound transformations in the era of AI to offer new perspectives for exploration beyond linguistic barriers. Although FLE training is focused on French, it also encompasses literary works from diverse cultural backgrounds, often inaccessible due to language barriers. In this context of "Digital Humanities," comparative literature is being reimagined through the lens of AI models, which raise questions about the role of the author and the use of AI as a tool for analysis in the literary field. From the above, we formulate the following research question: to what extent can generative AI facilitate access to foreign literatures for interactive analysis? How can Neural Machine Translation (NMT) and Large Language Models (LLMs) be effectively utilized to explore multilingual literary texts?