Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

WP

Benchmarks

Task NameDataset NameSOTA ResultTrend
LLM-generated text detectionWP-s Claude3 Opus
TPR @ FPR 1%96
18
LLM-generated text detectionWP-s GPT4 Turbo
TPR @ FPR 1%98.7
18
LLM-generated text detectionWP-s GPT4
TPR @ FPR 1%82
18
Inductive Knowledge Hypergraph Link PredictionWP 100
MRR0.222
15
Inductive Knowledge Hypergraph Link PredictionWP 75
MRR0.192
15
Inductive Knowledge Hypergraph Link PredictionWP 50
MRR0.212
15
Inductive Knowledge Hypergraph Link PredictionWP 25
MRR0.143
15
Node-Relation Inductive Link PredictionWP 100% unseen relations
MRR0.222
15
Node-Relation Inductive Link PredictionWP 75% unseen relations
MRR0.194
15
Node-Relation Inductive Link PredictionWP 50% unseen relations
MRR0.212
15
Node-Relation Inductive Link PredictionWP 25% unseen relations
MRR0.169
15
Robotic GrindingWP-E1 1.0 (test)
Execution Time (min)8
3
Robotic GrindingWP-E2
Execution Time (s)192.3
3
Robotic GrindingWP-E1
Execution Time (s)182.6
3
GrindingWP S5
In-limit Force Ratio (%)100
3
GrindingWP S2
In-limit Force Ratio100
3
GrindingWP-S1
Execution Time (s)5.3
3
Showing 17 of 17 rows