Reading Comprehension on BioASQ MRQA out-of-domain
67.62EMPrompt tuning
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Prompt tuningBackbone=Llama-3.2-3B-Base, Context=with context2026.02 | 67.62 | 81.5 | |
| Prompt tuningBackbone=Llama-3.2-1B-Base, Context=with context2026.02 | 65.96 | 78.8 | |
| ComprExITBackbone=Llama-3.2-3B-Base2026.02 | 59.64 | 73.87 | |
| ICAEBackbone=Llama-3.2-3B-Base2026.02 | 55.85 | 67.21 | |
| BeaconBackbone=Llama-3.2-3B-Base2026.02 | 54.52 | 66.86 | |
| ComprExITBackbone=Llama-3.2-1B-Base2026.02 | 48.4 | 63.74 | |
| 500xBackbone=Llama-3.2-3B-Base2026.02 | 47.67 | 60.82 | |
| Zero-shotBackbone=Llama-3.2-3B-Base, Context=with context2026.02 | 43.22 | 64.31 | |
| ICAEBackbone=Llama-3.2-1B-Base2026.02 | 42.62 | 54.09 | |
| BeaconBackbone=Llama-3.2-1B-Base2026.02 | 32.25 | 47.84 | |
| 500xBackbone=Llama-3.2-1B-Base2026.02 | 22.54 | 35.52 | |
| Zero-shotBackbone=Llama-3.2-3B-Base, Context=without context2026.02 | 20.94 | 25.76 | |
| Zero-shotBackbone=Llama-3.2-1B-Base, Context=without context2026.02 | 14.76 | 20.56 | |
| Zero-shotBackbone=Llama-3.2-1B-Base, Context=with context2026.02 | 6.98 | 28.84 |