Fig. 11From: Sequence-to-sequence modeling for graph representation learningType of substructure (solid: WL0, white hatch: WL1, black hatch:WL2)Back to article page