General-purpose Large Language Models outperform fine-tuned medical Large Language Models in medical hallucination detection tasks, according to the evaluation conducted by the authors of the MedHallu benchmark.

Referenced by nodes (2)

• Large Language Models concept
• MedHallu concept