On the MedMCQA benchmark, AMG-RAG achieves an accuracy of 66.34%, outperforming Meditron-70B (66.0%), Codex 5-shot CoT (59.7%), VOD (58.3%), Flan-PaLM (57.6%), PaLM (54.5%), GAL (120B, 52.9%), PubmedBERT (40.0%), SciBERT (39.0%), BioBERT (38.0%), and BERT (35.0%).

Authors

• Person: Not available Organization: arXiv
  Bridging the Gap Between LLMs and Evolving Medical Knowledge

Sources

• Bridging the Gap Between LLMs and Evolving Medical Knowledge arxiv.org arXiv via serper

Referenced by nodes (3)

• AMG-RAG concept
• BERT concept
• MEDMCQA concept