Wiki
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Wiki
1.139Faithfulness (KL distance)
11
Wiki (3% labeling ratio)
1.49Explainability Performance
11
Wiki (2% labeling ratio)
1.288Explainability Performance
11
Wiki
0.979Claude 3.5 Rewrite AUC
11
Wiki ChatGPT-generated (test)
98.79AUROC
11
Wiki (test)
58.9Accuracy
11
WIKI (test)
72.7AUC-PR
10
Wiki
1,786Estimated Cluster Count
10
Wiki 3% labeling ratio
17.8Accuracy
10
Wiki 2% labeling ratio
16.1Accuracy
10
Wiki20K (test)
0.246NPMI
10
Wiki-20m (test)
91.4AUC
10
Wiki (test)
0.208MAE
9
2Wiki
0.89Average Score
9
WIKI 1-shot (test)
0.277MRR
9
Wiki
2.97Rating
8
Wiki
45Purity (P1)
8
Wiki (dev)
15.77Perplexity (PPL)
8
Wiki 20% (train val)
51.2Accuracy
8
Wiki 5% (train/val)
33.7Accuracy
8
WIKI public (test)
0.46MRR
8
Wiki
0.49Purity
8
Wiki (test)
39.52P@1
8
Wiki (test)
73.34EM
8
WikiSmall
5.47FKGL
8