Axel Pichler🎉 Excited to share our (@janispagel, @nilsreiter) latest work at LaTeCH-CLfL 2025:
Prompt Engineering in Computational Literary Studies – Promise vs. Practice
We test how reliably LLMs generate sequence labels in a CLS context:
🔁 same prompts, different splits
🧠 prompt rewording
📎 performance-boosting phrases
Findings: results vary by split & wording; fixed phrases rarely help.
Evaluating LLM-Prompting for Sequence Labeling Tasks in Computational Literary Studies
Axel Pichler, Janis Pagel, Nils Reiter. Proceedings of the 9th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature (LaTeCH-CLfL 2025). 2025.