The 'LLM-as-Judge' evaluation method offers a closer alignment with human judgments of factual correctness compared to ROUGE, as validated by the human study conducted by the authors of 'Re-evaluating Hallucination Detection in LLMs'.

Authors

• Person: Not available Organization: arXiv
  Re-evaluating Hallucination Detection in LLMs - arXiv

Sources

• Re-evaluating Hallucination Detection in LLMs - arXiv arxiv.org arXiv via serper

Referenced by nodes (3)

• factual correctness concept
• ROUGE concept
• LLM-as-a-judge concept