Airline
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Airline
4.946NLL
22
Airline (test)
-0.702NLL
14
Airline (test)
39RMSE
10
Airline N=10000
0.89MSE
10
Airline N=5929413
0.763MSE
8
Airline N=1000000
0.77MSE
8
Airline (N=100000)
0.78MSE
8
Airline
37.8RMSE
7
Airline 200k subsample (test)
1.369NLL
7
airline 200k (test)
0.949RMSE
7
Airline
81.6Macro F1
6
τ³-airline 20-task (held-out)
60.2Pass@1
4
Airline (test)
0.346RMSE
4
AIRLINE n ≈ 10^6
—Primary metric
0