Table 1.

Spearman’s correlation coefficient between benchmark metrics for results using the masked graph model on the QM9 dataset.

	Validity	Uniqueness	Novelty	KL Div	Fréchet Dist
Validity	1.00	−0.56	−0.83	0.73	0.75
Uniqueness	−0.56	1.00	0.50	−0.32	−0.37
Novelty	−0.83	0.50	1.00	−0.94	−0.95
KL Div	0.73	−0.32	−0.94	1.00	0.99
Fréchet Dist	0.75	−0.37	−0.95	0.99	1.00