A Benchmark for Large Language Model Question Answering in Hematopoietic Cell and Bone Marrow Transplantation

Key Points

• The benchmark shows significant improvements in question answering accuracy within hematopoietic cell transplantation.
• Achieving over 80% accuracy in queries related to transplantation processes highlights the model's effectiveness.
• Analysis involves testing the model's responses against clinical queries provided by experts in the field.
• These findings indicate that large language models could support healthcare professionals in decision-making processes.