Do LLMs Really Struggle at NL-FOL Translation? Revealing their Strengths via a Novel Benchmarking Strategy
Paper • 2511.11816 • Published
Neurosymbolic AI, formal verification, interpretability, auto-formalization, learning specification from data, logic. Official repository of the Data Science and Automatic Verification Laboratory @ University of Udine
Fixing FOLIO and MALLS: Verified Annotations and an LLM-assisted Framework to Focus Human Relabeling