Papers › MetaGen Blended RAG: Higher Accuracy for Domain-Specific Q&A Without Fine-Tuning

MetaGen Blended RAG: Higher Accuracy for Domain-Specific Q&A Without Fine-Tuning

23 May 2025arXiv:2505.18247archive 2025-07-28

Kunal Sawarkar, Shivam R. Solanki, Abhilasha Mangal

Despite the widespread exploration of Retrieval-Augmented Generation (RAG), its deployment in enterprises for domain-specific datasets remains limited due to poor answer accuracy. These corpora, often shielded behind firewalls in private enterprise knowledge bases, having complex, domain-specific terminology, rarely seen by LLMs during pre-training; exhibit significant semantic variability across domains (like networking, military, or legal, etc.), or even within a single domain like medicine, and thus result in poor context precision for RAG systems. Currently, in such situations, fine-tuning or RAG with fine-tuning is attempted, but these approaches are slow, expensive, and lack generalization for accuracy as the new domain-specific data emerges. We propose an approach for Enterprise Search that focuses on enhancing the retriever for a domain-specific corpus through hybrid query indexes and metadata enrichment. This 'MetaGen Blended RAG' method constructs a metadata generation pipeline using key concepts, topics, and acronyms, and then creates a metadata-enriched hybrid index with boosted search queries. This approach avoids overfitting and generalizes effectively across domains. On the PubMedQA benchmark for the biomedical domain, the proposed method achieves 82% retrieval accuracy and 77% RAG accuracy, surpassing all previous RAG accuracy results without fine-tuning and sets a new benchmark for zero-shot results while outperforming much larger models like GPT3.5. The results are even comparable to the best fine-tuned models on this dataset, and we further demonstrate the robustness and scalability of the approach by evaluating it on other Q&A datasets like SQuAD, NQ etc.

PaperPDFCode

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Few-Shot LearningQuestion AnsweringRAGRetrievalRetrieval-augmented Generation

Datasets

Introduced by this paper, per the archive.

PubMedQA corpus with metadata

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Few-Shot Learning PubMedQA MetaGen Blended RAG (zero-shot) Accuracy 77.9 #1 of 2 Archive leaderboard report
Question Answering PubMedQA MetaGen Blended RAG (zero-shot) Accuracy 77.9 #9 of 30 Archive leaderboard report
RAG PubMedQA corpus with metadata MetaGen Blended RAG ANS-EM 77.90 #1 of 1 Archive leaderboard report
Retrieval PubMedQA MetaGen Blended RAG Accuracy (Top-1) 82.1 #1 of 1 Archive leaderboard report
Retrieval PubMedQA corpus with metadata MetaGen Blended RAG Accuracy (Top-1) 82.1 #1 of 1 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

RAG

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections