HyperAI

Open Domain Dialog On Kilt Wizard Of

Metrics

F1
KILT-F1
KILT-RL
R-Prec
ROUGE-L
Recall@5

Results

Performance results of various models on this benchmark

Comparison Table
Model NameF1KILT-F1KILT-RLR-PrecROUGE-LRecall@5
Model 10.00.00.064.790.082.15
Model 214.820.00.00.013.350.0
Model 317.30.00.00.015.930.0
Model 40.00.00.062.880.077.74
Model 53.092.182.0455.712.9275.59
Model 618.5711.7910.3655.3716.3678.45
Model 717.280.00.00.015.710.0
Model 818.3411.6310.4557.5516.6578.96
Model 90.00.00.059.110.069.1
Model 1019.1913.3911.9256.0817.0674.27
Model 1115.667.576.5541.5413.9468.25
Model 1212.860.00.00.011.770.0
kilt-a-benchmark-for-knowledge-intensive13.530.00.00.012.40.0
Model 1415.126.965.9141.0613.2767.13
Model 1512.154.84.4139.0611.4251.63
re2g-retrieve-rerank-generate-218.912.9811.3960.116.7679.98
Model 1713.750.00.00.012.810.0
Model 1811.852.21.8518.3510.1118.35
Model 1913.118.757.5957.7511.5774.61
Model 2015.194.373.7125.4613.2351.19
Model 210.00.00.041.060.067.13