HyperAI

Open Domain Question Answering On Kilt

Metrics

EM
F1
KILT-EM
KILT-F1
R-Prec
Recall@5

Results

Performance results of various models on this benchmark

Comparison Table
Model NameEMF1KILT-EMKILT-F1R-PrecRecall@5
Model 139.7548.4329.0934.759.4268.24
Model 246.0556.570.00.00.00.0
Model 345.2253.3836.3641.8363.7170.17
Model 453.7462.2438.7844.463.1668.19
Model 50.00.00.00.059.4268.24
Model 641.2749.5430.0634.7254.2965.52
Model 738.6447.0931.9937.5860.6646.79
Model 80.353.720.00.00.00.0
re2g-retrieve-rerank-generate-251.7360.9743.5649.870.7876.63
Model 1021.7528.690.00.00.00.0
Model 1151.5960.8335.3240.7359.8371.17
Model 120.00.00.00.062.664.95
kilt-a-benchmark-for-knowledge-intensive19.627.730.00.00.00.0
Model 140.00.00.00.060.3261.21
Model 150.00.00.00.060.2561.36
Model 1644.3952.3532.6937.9159.4967.06