Axel Pichler (German Studies, University of Vienna)

Lecture Series: Introduction to Digital Humanities

The field of Digital Humanities is increasingly exploring the potential and limitations of Large Language Models (LLMs) for domain-specific scholarly applications. In this talk, I will outline both the possibilities and challenges of using this technology in Literary Studies. My focus will be on approaches that assess LLM-generated results by drawing on literary expertise. I will discuss two specific use cases: first, classification tasks in which a literary concept is assigned to a text or text segment; and second, the computational generation of text interpretations—an area that has so far been considered “hardly plausible” (Gius 2019) and remains largely unexplored. Using an example from poetry interpretation, I will show how interpretations generated by LLMs can be evaluated from a more traditional literary studies perspective. My goal is to provide an initial overview of the extent to which LLMs can currently serve as useful tools in Digital Literary Studies.