Question Answering on SciQAG
93Accuracyfew-shot
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| few-shotModel=GPT-4.12025.06 | 93 | 0 | 0 | |
| ActIdxModel=Q-3B2025.06 | 67.9 | 20 | 23 | |
| ActIdx-FModel=Q-3B2025.06 | 67.7 | 3 | 3 | |
| PassIdxModel=Q-3B2025.06 | 67 | 1.1 | 1.6 | |
| ActIdxModel=Q-7B2025.06 | 66.6 | 32.6 | 33.6 | |
| ActIdx-BModel=Q-7B2025.06 | 66.5 | 30.8 | 32 | |
| InsOnlyModel=Q-7B2025.06 | 65.9 | 0.6 | 0.8 | |
| Repeat+Model=Q-7B2025.06 | 65.8 | 5.2 | 5.2 | |
| PassIdxModel=Q-7B2025.06 | 65.7 | 2.4 | 2.4 | |
| Repeat+Model=Q-3B2025.06 | 65.7 | 1.1 | 1.3 | |
| ActIdx-FModel=Q-7B2025.06 | 65.6 | 23.6 | 23.6 | |
| ActIdx-BModel=Q-3B2025.06 | 65.2 | 17.6 | 23.4 | |
| RepeatModel=Q-3B2025.06 | 64 | 0.9 | 1.2 | |
| RepeatModel=Q-7B2025.06 | 62.4 | 2.5 | 2.6 | |
| InsOnlyModel=Q-3B2025.06 | 61.2 | 0 | 0 |