FOLIO

The 'FOLIO' dataset/benchmark is used to evaluate the performance of language models in tasks that require generating outputs that are both syntactically and semantically correct.

Papers using FOLIO (3)

CRANE: Reasoning with constrained LLM generation2025Logic-LM: Empowering Large Language Models with Symbolic Solvers for Faithful Logical Reasoning2023 · 100 citesLLM-ARC: Enhancing LLMs with an Automated Reasoning Critic2024 · 2 cites