Question Answering on MRQA
62.58HotpotQA EMVanilla-RAG
Evaluation Results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Vanilla-RAGBackbone=Qwen-3-8B2026.06 | 62.58 | 78.81 | 66.28 | 75.05 | 38.77 | 60.22 | 55.1 | 70.99 | 73.63 | 81.07 | 72.29 | 84.32 | |
| Knowledgeable-r1Backbone=Qwen-3-8B2026.06 | 62.55 | 78.84 | 67 | 75.71 | 39.17 | 60.46 | 54.9 | 70.66 | 74.39 | 81.78 | 72.82 | 84.77 | |
| ShiftBackbone=Qwen-3-8B2026.06 | 62.48 | 79.08 | 68.02 | 76.27 | 39.39 | 60.93 | 57.38 | 73.45 | 75.23 | 82.28 | 73.46 | 85.5 | |
| SFTBackbone=Qwen-3-8B2026.06 | 59.79 | 75.82 | 70.72 | 78.12 | 40.88 | 60.93 | 53.96 | 70.36 | 75.02 | 82.14 | 74.25 | 84.78 | |
| CK-PLUGBackbone=Qwen-3-8B2026.06 | 52.36 | 67.9 | 60.93 | 69.26 | 35.23 | 55.59 | 48.89 | 65.56 | 70.8 | 78.09 | 67.62 | 80.39 | |
| ShiftBackbone=Qwen-3-0.6B2026.06 | 47.03 | 63.36 | 41.13 | 50.38 | 32.53 | 50.58 | 47.34 | 62.95 | 60.54 | 68.72 | 58.34 | 71.29 | |
| Knowledgeable-r1Backbone=Qwen-3-0.6B2026.06 | 44.48 | 60.29 | 44.61 | 53.43 | 26.54 | 40.89 | 41.04 | 53.25 | 52.28 | 58.7 | 50.89 | 61.9 | |
| Vanilla-RAGBackbone=Qwen-3-0.6B2026.06 | 43.64 | 59.73 | 27.57 | 35.66 | 22.77 | 36.28 | 40.41 | 53.29 | 46.55 | 53.99 | 46.96 | 59.14 | |
| CK-PLUGBackbone=Qwen-3-0.6B2026.06 | 41.94 | 58.31 | 28.17 | 36.45 | 22.98 | 43.12 | 41.59 | 57.86 | 48.61 | 58.93 | 46.26 | 63.3 | |
| SFTBackbone=Qwen-3-0.6B2026.06 | 39.57 | 54.6 | 41.81 | 50.16 | 26.45 | 43.18 | 38.25 | 52.7 | 60.67 | 67.74 | 55.8 | 68.67 | |
| CtrlABackbone=Qwen-3-8B2026.06 | 34.13 | 47.28 | 52.9 | 61.01 | 26.88 | 43.52 | 36.34 | 52.55 | 59.68 | 66.41 | 45.65 | 59.05 | |
| CtrlABackbone=Qwen-3-0.6B2026.06 | 20.88 | 31.48 | 10.28 | 13.5 | 10.87 | 19.14 | 21.4 | 30.24 | 23.11 | 28.69 | 24.77 | 34.3 | |
| No-RAGBackbone=Qwen-3-8B2026.06 | 11.9 | 21.53 | 42.64 | 49.76 | 2.14 | 6.22 | 15.36 | 28.07 | 50.31 | 56.52 | 12.08 | 23.7 | |
| No-RAGBackbone=Qwen-3-0.6B2026.06 | 2.27 | 7.51 | 3.49 | 5.33 | 0.52 | 3.32 | 2.27 | 6.76 | 8.2 | 12.27 | 2.1 | 8.77 |