LPDS: Evaluating LLM Robustness Through Logic-Preserving Difficulty Scaling

Recommended citation: Philipp Mondorf, Samuel J. Bell, Jesse Dodge, and Dieuwke Hupkes. 2026. LPDS: Evaluating LLM Robustness Through Logic-Preserving Difficulty Scaling. In Proceedings of the Third Conference on Language Modeling. URL: https://openreview.net/forum?id=wAR7NJB3FV.
Download Paper