Open-domain Question Answering on 2WikiMultiHopQA in-domain
62.6F1 ScoreUR2
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| UR2Backbone=Qwen-2.5-7B, Type=Our Implementations2025.08 | 62.6 | 65.2 | |
| UR2Model=Qwen-2.5-7B, Type=Our Implementations2025.08 | 62.6 | 65.2 | |
| R1-Searcher++Backbone=Qwen-2.5-7B, Type=RAG-RL2025.08 | 61.2 | 64.4 | |
| Vanilla RLBackbone=Qwen-2.5-7B, Type=Our Implementations2025.08 | 61.2 | 62.4 | |
| Vanilla RLBackbone=LLaMA-3.1-8B, Type=Our Implementations2025.08 | 61.2 | 64.2 | |
| R1-Searcher++Model=Qwen-2.5-7B, Type=RAG-RL2025.08 | 61.2 | 64.4 | |
| Vanilla RLModel=Qwen-2.5-7B, Type=Our Implementations2025.08 | 61.2 | 62.4 | |
| Vanilla RLModel=LLaMA-3.1-8B, Type=Our Implementations2025.08 | 61.2 | 64.2 | |
| Search-R1Backbone=Qwen-2.5-7B, Type=RAG-RL2025.08 | 61 | 63.8 | |
| Search-R1Model=Qwen-2.5-7B, Type=RAG-RL2025.08 | 61 | 63.8 | |
| UR2Backbone=LLaMA-3.1-8B, Type=Our Implementations2025.08 | 60.1 | 63.2 | |
| UR2Model=LLaMA-3.1-8B, Type=Our Implementations2025.08 | 60.1 | 63.2 | |
| R1-SearcherBackbone=LLaMA-3.1-8B, Type=RAG-RL2025.08 | 59.6 | 62.2 | |
| R1-SearcherModel=LLaMA-3.1-8B, Type=RAG-RL2025.08 | 59.6 | 62.2 | |
| R1-SearcherBackbone=Qwen-2.5-7B, Type=RAG-RL2025.08 | 57.9 | 63.6 | |
| R1-SearcherModel=Qwen-2.5-7B, Type=RAG-RL2025.08 | 57.9 | 63.6 | |
| UR2Backbone=Qwen-2.5-3B, Type=Our Implementations2025.08 | 55.2 | 58.6 | |
| UR2Model=Qwen-2.5-3B, Type=Our Implementations2025.08 | 55.2 | 58.6 | |
| Vanilla RLBackbone=Qwen-2.5-3B, Type=Our Implementations2025.08 | 54.9 | 58 | |
| Vanilla RLModel=Qwen-2.5-3B, Type=Our Implementations2025.08 | 54.9 | 58 | |
| Self-AskBackbone=GPT-4.1-mini, Type=Advanced RAG2025.08 | 52.7 | 57.4 | |
| Search-R1Backbone=Qwen-2.5-3B, Type=RAG-RL2025.08 | 49.5 | 53.4 | |
| Search-R1Model=Qwen-2.5-3B, Type=RAG-RL2025.08 | 49.5 | 53.4 | |
| CoTBackbone=GPT-4.1-mini, Type=Vanilla Methods2025.08 | 48.6 | 60.8 | |
| Search-o1Model=GPT-4o-mini, Type=Advanced RAG2025.08 | 47.3 | 52 | |
| RATBackbone=GPT-4.1-mini, Type=Advanced RAG2025.08 | 45.7 | 49 | |
| Search-o1Model=Qwen-2.5-7B, Type=Advanced RAG2025.08 | 45.2 | 48.6 | |
| Search-o1Backbone=GPT-4.1-mini, Type=Advanced RAG2025.08 | 44.4 | 60.6 | |
| Standard RAGModel=GPT-4o-mini, Type=Vanilla Methods2025.08 | 43 | 45.8 | |
| Standard RAGBackbone=GPT-4.1-mini, Type=Vanilla Methods2025.08 | 41.3 | 52.4 | |
| ZeroSearchBackbone=Qwen-2.5-7B, Type=RAG-RL2025.08 | 38.4 | 38.6 | |
| ZeroSearchModel=Qwen-2.5-7B, Type=RAG-RL2025.08 | 38.4 | 38.6 | |
| Search-o1Model=LLaMA-3.1-8B, Type=Advanced RAG2025.08 | 37.5 | 38.4 | |
| Self-AskModel=GPT-4o-mini, Type=Advanced RAG2025.08 | 36.9 | 40 | |
| CoTModel=GPT-4o-mini, Type=Vanilla Methods2025.08 | 35 | 35.4 | |
| RATModel=GPT-4o-mini, Type=Advanced RAG2025.08 | 34.1 | 34.8 | |
| Standard RAGBackbone=Qwen-2.5-7B, Type=Vanilla Methods2025.08 | 32.8 | 37.6 | |
| Standard RAGModel=Qwen-2.5-7B, Type=Vanilla Methods2025.08 | 32.8 | 37.6 | |
| Search-o1Model=Qwen-2.5-3B, Type=Advanced RAG2025.08 | 30.8 | 31.8 | |
| Standard RAGBackbone=Qwen-2.5-3B, Type=Vanilla Methods2025.08 | 29.8 | 30.4 | |
| Standard RAGModel=Qwen-2.5-3B, Type=Vanilla Methods2025.08 | 29.8 | 30.4 | |
| Self-AskModel=LLaMA-3.1-8B, Type=Advanced RAG2025.08 | 27.3 | 29.8 | |
| Standard RAGBackbone=LLaMA-3.1-8B, Type=Vanilla Methods2025.08 | 26.2 | 26.4 | |
| Standard RAGModel=LLaMA-3.1-8B, Type=Vanilla Methods2025.08 | 26.2 | 26.4 | |
| Zero-SearchBackbone=Qwen-2.5-3B, Type=RAG-RL2025.08 | 26.1 | 27.6 | |
| Zero-SearchModel=Qwen-2.5-3B, Type=RAG-RL2025.08 | 26.1 | 27.6 | |
| CoTBackbone=Qwen-2.5-7B, Type=Vanilla Methods2025.08 | 25.1 | 27.6 | |
| CoTModel=Qwen-2.5-7B, Type=Vanilla Methods2025.08 | 25.1 | 27.6 | |
| RATModel=Qwen-2.5-7B, Type=Advanced RAG2025.08 | 23.3 | 23.6 | |
| CoTBackbone=Qwen-2.5-3B, Type=Vanilla Methods2025.08 | 22.7 | 22.6 | |
| CoTModel=Qwen-2.5-3B, Type=Vanilla Methods2025.08 | 22.7 | 22.6 | |
| Self-AskModel=Qwen-2.5-7B, Type=Advanced RAG2025.08 | 22.2 | 45.4 | |
| Self-AskModel=Qwen-2.5-3B, Type=Advanced RAG2025.08 | 21 | 28.8 | |
| CoTBackbone=LLaMA-3.1-8B, Type=Vanilla Methods2025.08 | 16.4 | 17.8 | |
| CoTModel=LLaMA-3.1-8B, Type=Vanilla Methods2025.08 | 16.4 | 17.8 | |
| RATModel=LLaMA-3.1-8B, Type=Advanced RAG2025.08 | 16.4 | 15.6 | |
| RATModel=Qwen-2.5-3B, Type=Advanced RAG2025.08 | 15.1 | 15.4 |