Multi-hop Question Answering on Musique fixed split (test)
39.84Exact Match (EM)Graph-AdaptR1
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Graph-AdaptR1Backbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Graph-based2026.05 | 39.84 | 53.42 | |
| Graph-R1Backbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Graph-based2026.05 | 37.5 | 48.33 | |
| Search-AdaptR1Backbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Chunk-based2026.05 | 25 | 34.31 | |
| Search-R1Backbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Chunk-based2026.05 | 17.18 | 21.45 | |
| R1-SearcherBackbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Chunk-based2026.05 | 10.16 | 16.63 | |
| StandardRAGBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Chunk-based2026.05 | 9.38 | 17.31 | |
| GraphRAGBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Graph-based2026.05 | 7.03 | 15.14 | |
| R1Backbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=None2026.05 | 7.03 | 14.53 | |
| HippoRAG2Backbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Graph-based2026.05 | 6.25 | 12.37 | |
| HyperGraphRAGBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Graph-based2026.05 | 6.25 | 20.4 | |
| SFTBackbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Chunk-based2026.05 | 5.47 | 10.02 | |
| LightRAGBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Graph-based2026.05 | 3.91 | 14.39 | |
| NaiveGenerationBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=None2026.05 | 3.13 | 11.45 | |
| PathRAGBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Graph-based2026.05 | 3.13 | 11.49 | |
| StandardRAGBackbone=Qwen2.5-7B-Instruct, Interaction Type=Prompt Engineering, Knowledge Representation=Chunk-based2026.05 | 0.78 | 4.53 | |
| NaiveGenerationBackbone=Qwen2.5-7B-Instruct, Interaction Type=Prompt Engineering, Knowledge Representation=None2026.05 | 0 | 4.06 |