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

HIV

Benchmarks

Task NameDataset NameSOTA ResultTrend
Graph ClassificationHIV
ROC-AUC0.832
160
Molecular property predictionHIV (scaffold split)
ROC-AUC79.9
56
Graph ClassificationHIV
Accuracy62.18
52
Molecular property predictionHIV
ROC AUC79.88
51
Molecular predictionHIV (test)
ROC AUC79.42
46
Molecular ClassificationHIV
ROC-AUC85.6
42
Molecular property predictionHIV 50-shot (test)
ROC-AUC68.19
40
Molecule Property PredictionHIV MoleculeNet (test)
AUROC84.81
40
Policy LearningHIV dataset
Estimated Expected Outcome6.063
36
Graph classificationHIV OGB (test)
ROC AUC80.67
33
Graph ClassificationHIV MoleculeNet (test)
AUPRC35.47
27
Graph ClassificationHIV GraphOOD (test)
ROC-AUC73.31
26
Molecular predictionHIV full-shot (test)
ROC-AUC79.42
25
Molecular property classificationHIV (scaffold)
ROC AUC0.789
25
Surrogate detection under model extraction attackHIV
AUC100
20
ClassificationHIV random 6:2:2 (test)
ROC-AUC0.7833
17
Graph-level molecular classificationHIV
MCC0.533
16
Molecular predictionHIV 100-shot (test)
ROC AUC0.6979
15
Molecular Property PredictionHIV MoleculeNet
ROC-AUC77.88
14
Graph ClassificationHIV Molecule OGB (standard)
ROC AUC0.7784
14
Hit IdentificationHIV-Hi Lo-Hi benchmark
AUPRC15.8
13
Property PredictionHIV
ROC-AUC80.8
13
Drug-Drug Interaction PredictionHIV
F1 Score95
12
Graph ClassificationHIV MoleculeNet scaffold-split (test)
AUROC (%)80.8
12
Covariate Shift CalibrationHIV
Mean Absolute Deviation (MAD)0.013
12
Showing 25 of 56 rows