Multi-hop Question Answering on HotpotQA v1.0 (test)
2.28Latency (s)No Retrieval
Evaluation Results
| Method | Links | |
|---|---|---|
| No RetrievalBackbone=Llama3.1-8B2025.03 | 2.28 | |
| VanillaRAGBackbone=Llama3.1-8B, k=42025.03 | 2.64 | |
| VanillaRAGBackbone=Llama3.1-8B, k=82025.03 | 2.73 | |
| L-RAGBackbone=Llama3.1-8B, k=42025.03 | 2.8 | |
| L-RAGBackbone=Llama3.1-8B, k=82025.03 | 2.92 | |
| HyDEBackbone=Llama3.1-8B2025.03 | 5.28 | |
| IRCoTBackbone=Llama3.1-8B2025.03 | 17.3 | |
| SelfAskBackbone=Llama3.1-8B2025.03 | 19.59 | |
| IterRetGenBackbone=Llama3.1-8B2025.03 | 42.36 |