Multi-hop Question Answering on MuSiQue first 1000 samples (dev)
33.5EMPathfinder
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| PathfinderBase Model=Gemma-2-9B-Instruct, Distribution Setting=in-distribution2025.12 | 33.5 | 38.1 | 42.41 | |
| DeepRAG (Imi-retrieve only)Base Model=Qwen-2.5-7B-Instruct, Distribution Setting=in-distribution2025.12 | 32.5 | 35 | 40.53 | |
| PathfinderBase Model=Qwen-2.5-7B-Instruct, Distribution Setting=in-distribution2025.12 | 32.1 | 35.2 | 40.86 | |
| DeepRAG (Imi-retrieve only)Base Model=Gemma-2-9B-Instruct, Distribution Setting=in-distribution2025.12 | 31.4 | 34.3 | 40.68 | |
| DeepRAG (Imi)Base Model=Gemma-2-9B-Instruct, Distribution Setting=in-distribution2025.12 | 28.3 | 31.1 | 36.22 | |
| DeepRAG (Imi)Base Model=Qwen-2.5-7B-Instruct, Distribution Setting=in-distribution2025.12 | 18.4 | 20.8 | 25.37 |