Simple questions used in benchmarks by Bordes et al. (2015) and Joshi et al. (2017) are typically short, open-ended queries with a single, verifiable answer that require large language models to draw on internalized representations but fail to capture multiple elements of deeper knowledge.

Authors

• Person: Not available Organization: arXiv
  A Knowledge Graph-Based Hallucination Benchmark for Evaluating ...

Sources

• A Knowledge Graph-Based Hallucination Benchmark for Evaluating ... arxiv.org arXiv via serper

Referenced by nodes (1)

• Large Language Models concept