Multi-hop Question Answering on 2WikiMultiHop
73.4SubEMEHRAG
Evaluation results
| Method | Links | ||
|---|---|---|---|
| EHRAGGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 73.4 | 70.6 | |
| LinearRAGGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 70.2 | 63.7 | |
| GFM-RAGGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 66.8 | 59.6 | |
| HippoRAGGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 66.1 | 59.9 | |
| HippoRAG2Generation Strategy=Retrieval-Augmented-Generation Methods2026.04 | 62.7 | 55 | |
| LightRAGGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 55.2 | 39 | |
| E2GraphRAGGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 54.3 | 38.1 | |
| RAPTORGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 50.1 | 42.1 | |
| GraphRAGGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 49.4 | 41.6 | |
| Vanilla RAGGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 48.6 | 43 | |
| G-retrieverGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 46.6 | 27.1 | |
| GPT-4o-miniGeneration Strategy=Direct Zero-shot LLM Inference2026.04 | 36.3 | 31.4 | |
| llama-8BGeneration Strategy=Direct Zero-shot LLM Inference2026.04 | 33.6 | 16.2 | |
| KGPGeneration Strategy=Retrieval-Augmented-Generation Methods2026.04 | 31.6 | 30 | |
| Qwen3-8BGeneration Strategy=Direct Zero-shot LLM Inference2026.04 | 29.8 | 27.1 | |
| GPT-3.5-turboGeneration Strategy=Direct Zero-shot LLM Inference2026.04 | 28.7 | 31 | |
| llama-13BGeneration Strategy=Direct Zero-shot LLM Inference2026.04 | 21.9 | 10.5 |