Reading Comprehension On Reclor
Metriken
Test
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
Modellname | Test |
---|---|
Modell 1 | 69.7 |
Modell 2 | 56.9 |
Modell 3 | 49.8 |
Modell 4 | 77.2 |
Modell 5 | 25.3 |
Modell 6 | 40.5 |
Modell 7 | 63.2 |
Modell 8 | 50.3 |
Modell 9 | 59.2 |
Modell 10 | 79.3 |
Modell 11 | 79.2 |
logiformer-a-two-branch-graph-transformer | 63.5 |
Modell 13 | 62.6 |
Modell 14 | 69.3 |
fact-driven-logical-reasoning | 58.9 |
reclor-a-reading-comprehension-dataset-1 | 50.4 |
Modell 17 | 26.0 |
Modell 18 | 61.7 |
answer-uncertainty-and-unanswerability-in-1 | 71.0 |
Modell 20 | 66.7 |
idol-indicator-oriented-logic-pre-training | 80.6 |
Modell 22 | 80.0 |
Modell 23 | 61.7 |
Modell 24 | 65.6 |
reclor-a-reading-comprehension-dataset-1 | 47.3 |
Modell 26 | 60.2 |
Modell 27 | 57.3 |
Modell 28 | 55.6 |
merit-meta-path-guided-contrastive-learning | 79.3 |
reclor-a-reading-comprehension-dataset-1 | 48.5 |
Modell 31 | 60.3 |
logic-driven-context-extension-and-data | 76.1 |
Modell 33 | 57.7 |
Modell 34 | 68.6 |
Modell 35 | 60.2 |
Modell 36 | 57.7 |
Modell 37 | 56.0 |
dagn-discourse-aware-graph-network-for | 58.2 |
reclor-a-reading-comprehension-dataset-1 | 56.0 |