Paper proposes LLM-assisted relabeling to fix FOLIO, MALLS benchmarks

Arxiv CS.CL (NLP/language models)

The paper introduces verified annotations and an LLM-assisted framework to focus human relabeling for NL-to-FOL translation benchmarks. The approach aims to improve accuracy in neurosymbolic AI and NLI tasks.

neuro-symbolic

Arxiv CS.CL (NLP/language models)·Andrea BrunelloCristian CurabaLuca GeattiMichele MignaniAngelo MontanariNicola Saccomanno

8 days ago