Wikipedia
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Wikipedia European Languages
96.7nDCG@10
8
Wikipedia All Languages
96.3nDCG@10
8
Wikipedia
-0.13Fidelity
8
Wikipedia
99.31BLEU-4
8
Wikipedia
0.056CRPSsum
8
Wikipedia
75.392COMET Score
8
Wikipedia GluonTS (test)
0.944MAE
8
Wikipedia (test)
1.28MSE
8
Wikipedia converted link prediction (test)
57.3NDCG@10
8
Wikipedia converted link prediction (val)
0.564NDCG@10
8
Wikipedia Gen: Polish 4o-mini, Regen: Para 4o-mini 1.0 (test)
0.815AUROC
8
Wikipedia Gen: Revise 4.1, Regen: Para 4o-mini 1.0 (test)
0.853AUROC
8
Wikipedia Gen: Revise 4o-mini, Regen: Para 4o-mini 1.0 (test)
88.76AUROC
8
Wikipedia Gen: Para Phi, Regen: Para 4o-mini 1.0 (test)
0.87AUROC
8
Wikipedia Gen: Para Grok, Regen: Para 4o-mini 1.0 (test)
86.46AUROC
8
Wikipedia Gen: Para DeepSeek, Regen: Para 4o-mini 1.0 (test)
0.865AUROC
8
Wikipedia Gen: Para 4.1, Regen: Para 4o-mini 1.0 (test)
87.5AUROC
8
Wikipedia Gen: Para 4o-mini, Regen: Para 4o-mini 1.0 (test)
0.907AUROC
8
Wikipedia (test)
0.134Literal F1
8
Wikipedia (test)
78.6MRR
8
Wikipedia (80% train ratio)
83.63Micro-F1
8
Wikipedia (70% train ratio)
83.52Micro F1
8
Wikipedia (60% train ratio)
83.37Micro-F1
8
Wikipedia 50% train ratio
83.16Micro-F1
8
Wikipedia (40% train ratio)
82.93Micro-F1 Score
8