TABLE 3. Model Comparison With Cutting Edge Approaches on the DAVIS Dataset.

Models	CI (std)	MSE	(std)	AUPC (std)
KronRLS [27]	0.869 (0.001)	0.379	.407 (0.005)	0.661 (0.010)
SimBoost [26]	0.873 (0.002)	0.282	0.644 (0.006)	0.709 (0.008)
String Representation Based Approaches
DeepDTA [29]	0.878 (0.004)	0.261	0.630 (0.017)	0.714 (0.010)
MT-DTI (wo-FT) [32]	0.875 (0.003)	0.268	0.633 (0.013)	0.700 (0.011)
MT-DTI [32]	0.887 (0.001)	0.245	0.665 (0.014)	0.730 (0.014)
DeepCPI [28]	0.867	0.293	0.607	0.705
WideDTA [7]	0.886 (0.003)	0.262	0.633 (0.011)	0.711 (0.012)
GANsDTA [13]	0.881 (0.005)	0.276	0.653 (0.015)	0.653 (0.017)
Attention-DTA [14]	0.887 (0.005)	0.245	0.657 (0.024)	0.746 (0.024)
Graph Representation Based Approaches
GAT [11]	0.892 (0.003)	0.232	0.662 (0.010)	0.728 (0.016)
GIN [11]	0.893 (0.003)	0.229	0.649 (0.013)	0.720 (0.016)
GIN [12]	0.899 (0.003)	0.220	0.623 (0.011)	0.726 (0.015)
DeepGS [30]	0.880 (0.005)	0.252	0.686 (0.012)	0.763 (0.012)
DeepH-DTA*	0.924 (0.001)	0.195	0.725 (0.009)	0.801 (0.010)

The * denote the proposed architecture