WP
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
WP-s Claude3 Opus
96TPR @ FPR 1%
18
WP-s GPT4 Turbo
98.7TPR @ FPR 1%
18
WP-s GPT4
82TPR @ FPR 1%
18
WP
93.35Accuracy
16
WP 100
0.222MRR
15
WP 75
0.192MRR
15
WP 50
0.212MRR
15
WP 25
0.143MRR
15
WP 100% unseen relations
0.222MRR
15
WP 75% unseen relations
0.194MRR
15
WP 50% unseen relations
0.212MRR
15
WP 25% unseen relations
0.169MRR
15
WP Domain Generalization (DG-MGT)
93.35Accuracy
12
WP-E1 1.0 (test)
8Execution Time (min)
3
WP-E2
192.3Execution Time (s)
3
WP-E1
182.6Execution Time (s)
3
WP S5
100In-limit Force Ratio (%)
3
WP S2
100In-limit Force Ratio
3
WP-S1
5.3Execution Time (s)
3