Question Answering on OpenBookQA (F1)
29.84F1 ScoreLIDAS
Evaluation Results
| Method | Links | |
|---|---|---|
| LIDASModel Size=1.7B, Cooldown Protocol=Standard, Zero-shot=true2025.12 | 29.84 | |
| BaselineModel Size=1.7B, Cooldown Protocol=Standard, Zero-shot=true2025.12 | 29.57 | |
| LN-ScalingModel Size=1.7B, Cooldown Protocol=Standard, Zero-shot=true2025.12 | 29.11 | |
| MIDASModel Size=1.7B, Cooldown Protocol=Standard, Zero-shot=true2025.12 | 28.8 | |
| LIDASModel Size=360M, Cooldown Protocol=Standard, Zero-shot=true2025.12 | 26.63 | |
| MIDASModel Size=360M, Cooldown Protocol=Standard, Zero-shot=true2025.12 | 24.57 | |
| LN-ScalingModel Size=360M, Cooldown Protocol=Standard, Zero-shot=true2025.12 | 23.14 | |
| BaselineModel Size=360M, Cooldown Protocol=Standard, Zero-shot=true2025.12 | 22.89 |