Question Answering on FEVER (test)
89.38AccuracyML Routing
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| ML RoutingAdversarial attack=TextAttack2026.06 | 89.38 | 76.25 | 81.11 | 13.23 | |
| LWV FusionAdversarial attack=TextAttack2026.06 | 77.35 | 76.43 | 70.09 | 29.91 | |
| Geometric-Based ModelAdversarial attack=TextAttack2026.06 | 76.25 | 77.25 | — | 33.2 | |
| FreeLBAdversarial attack=TextAttack2026.06 | 67.29 | 67.29 | — | 42.71 | |
| Uncertainty-Based ModelAdversarial attack=TextAttack2026.06 | 63.45 | 63.45 | 43 | 58 | |
| Entropy-Based ModelAdversarial attack=TextAttack2026.06 | 58.1 | 58.1 | 56.38 | 43.62 | |
| Base Model (no defence)Adversarial attack=TextAttack2026.06 | 24.46 | 36.2 | — | 75.5 |