Cross-domain generalization on FEVER (test)
99.7AccuracyLPF-SPN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| LPF-SPNSeed selection=Best seed per domain2026.03 | 99.7 | 0 | |
| LPF-LearnedSeed selection=Best seed per domain2026.03 | 99.7 | — | |
| VAE-OnlySeed selection=Best seed per domain2026.03 | 99.7 | — | |
| VAESeed selection=Best seed per domain2026.03 | 99.7 | — |