Retrieval-Augmented Generation on StrategyQA 100-query benchmark
69.8Mean ScoreRAGVUE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RAGVUEMetric=clarity2025.12 | 69.8 | 10 | |
| RAGASMetric=context_relevance2025.12 | 55 | 26.4 | |
| RAGASMetric=faithfulness2025.12 | 52.1 | 40.3 | |
| RAGVUEMetric=retrieval_coverage2025.12 | 50.3 | 27.9 | |
| RAGVUEMetric=retrieval_relevance2025.12 | 42 | 31.6 | |
| RAGVUEMetric=strict_faithfulness2025.12 | 40 | 49.2 | |
| RAGASMetric=response_groundedness2025.12 | 39 | 46 | |
| RAGVUEMetric=answer_relevance2025.12 | 37.2 | 25.5 | |
| RAGASMetric=answer_relevancy2025.12 | 24 | 30.7 | |
| RAGVUEMetric=answer_completeness2025.12 | 12.1 | 22.5 |