Firstly, we obtain the motif adjacency matrix by motif mining, which encodes the Motif higher order structure. Then we couple the motif adjacency matrix with the adjacency matrix of the original graph ...

Next, the fuzzy adjacency matrix and the fuzzy contextualized representation are concatenated to create the final representation and fed into GCN layers to capture the high-level features of the ...

During table generation process, the words are assigned unique IDs. During html-to-image conversion, these words are localized with bounding boxes and transformed on image transformation accordingly.