MetaGen Blended RAG (zero-shot)

Reported on 3 benchmarks across 3 tasks · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology2 results

Few-Shot LearningonPubMedQA
Accuracy· 2025-05-23
77.9
SOTA
MetaGen Blended RAG: Higher Accuracy for Domain-Specific Q&A Without Fine-Tuning arXiv:2505.18247
Meta-LearningonPubMedQA
Accuracy· 2025-05-23
77.9
SOTA
MetaGen Blended RAG: Higher Accuracy for Domain-Specific Q&A Without Fine-Tuning arXiv:2505.18247

Natural Language Processing1 result

Question AnsweringonPubMedQA
Accuracy· 2025-05-23
77.9
best: 81.6 (Meditron-70B (CoT + SC))
MetaGen Blended RAG: Higher Accuracy for Domain-Specific Q&A Without Fine-Tuning arXiv:2505.18247