Multi-hop Question Answering on 2WikiMultiHopQA first 1000 samples (dev)
75EMPathfinder
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| PathfinderBase Model=Qwen-2.5-7B-Instruct, Distribution Setting=in-distribution2025.12 | 75 | 76.3 | 78.38 | |
| PathfinderBase Model=Gemma-2-9B-Instruct, Distribution Setting=in-distribution2025.12 | 72.5 | 74.6 | 76.77 | |
| DeepRAG (Imi-retrieve only)Base Model=Gemma-2-9B-Instruct, Distribution Setting=in-distribution2025.12 | 71.9 | 73.8 | 75.46 | |
| DeepRAG (Imi-retrieve only)Base Model=Qwen-2.5-7B-Instruct, Distribution Setting=in-distribution2025.12 | 69.4 | 70.9 | 73.48 | |
| DeepRAG (Imi)Base Model=Gemma-2-9B-Instruct, Distribution Setting=in-distribution2025.12 | 58 | 59.7 | 62.06 | |
| DeepRAG (Imi)Base Model=Qwen-2.5-7B-Instruct, Distribution Setting=in-distribution2025.12 | 47.7 | 48.9 | 52.87 |