Single-hop Question Answering on NQ
43EMHELP
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| HELPBackbone=Qwen3-30B-A3B-Instruct-25072026.02 | 43 | 56.9 | |
| HippoRAG2Backbone=Qwen3-30B-A3B-Instruct-25072026.02 | 40.4 | 54.4 | |
| Llama3.3-70BReasoning Protocol=Direct Reasoning (w/o Retrieval), Parameter Count=70B2025.01 | 36 | 48.7 | |
| Search-o1Reasoning Protocol=Retrieval-augmented Reasoning with Reason-in-Documents, Parameter Count=32B2025.01 | 34 | 49.7 | |
| RAgent-QwQ-32BReasoning Protocol=Retrieval-augmented Reasoning, Parameter Count=32B2025.01 | 33.8 | 48.4 | |
| RAG-Qwen2.5-32BReasoning Protocol=Retrieval-augmented Reasoning, Parameter Count=32B2025.01 | 33.4 | 49.3 | |
| RAgent-Qwen2.5-32BReasoning Protocol=Retrieval-augmented Reasoning, Parameter Count=32B2025.01 | 32.4 | 47.8 | |
| LinearRAGBackbone=Qwen3-30B-A3B-Instruct-25072026.02 | 31.6 | 43.8 | |
| HyperGraphRAGBackbone=Qwen3-30B-A3B-Instruct-25072026.02 | 30.3 | 42.1 | |
| RAG-QwQ-32BReasoning Protocol=Retrieval-augmented Reasoning, Parameter Count=32B2025.01 | 29.6 | 44.4 | |
| Qwen2.5-72BReasoning Protocol=Direct Reasoning (w/o Retrieval), Parameter Count=72B2025.01 | 27.6 | 41.2 | |
| QwQ-32BReasoning Protocol=Direct Reasoning (w/o Retrieval), Parameter Count=32B2025.01 | 23 | 33.1 | |
| Qwen2.5-32BReasoning Protocol=Direct Reasoning (w/o Retrieval), Parameter Count=32B2025.01 | 22.8 | 33.9 |