AI hallucination benchmarks aim to quantify how often language models produce...

https://www.scribd.com/document/1013175958/When-Summaries-Lie-A-Case-study-of-Models-That-Summarize-Well-but-Fail-to-Admit-Ignorance-147755

AI hallucination benchmarks aim to quantify how often language models produce false or misleading information—an issue that directly affects trust and reliability in real-world applications