The examples.txt file shows several example sentences with the chunks extracted from the lexicon and the CCG supertagging and parsing results of our approach (i.e., BERT + A-GCN (Chunk)).

For each example, there are four groups of lines:

The first line is the sentence; the second line shows the chunks with their starting and ending position; the third groups of lines show the output supertags for each word; the last line shows the CCG parsing results.

Here is an example:

ID=2
The finger-pointing has already begun .
(2, 3): has already (4, 4): begun
The	NP[nb]/N
finger-pointing	N
has	(S[dcl]\NP)/(S[pt]\NP)
already	(S\NP)\(S\NP)
begun	S[pt]\NP
.	.
(<T S[dcl] 1 2> (<T NP[nb] 0 2> (<L NP[nb]/N POS POS The NP[nb]/N>) (<L N POS POS finger-pointing N>) ) (<T S[dcl]\NP 0 2> (<T (S[dcl]\NP)/(S[pt]\NP) 0 2> (<L (S[dcl]\NP)/(S[pt]\NP) POS POS has (S[dcl]\NP)/(S[pt]\NP)>) (<L (S\NP)\(S\NP) POS POS already (S\NP)\(S\NP)>) ) (<T S[pt]\NP 0 2> (<L S[pt]\NP POS POS begun S[pt]\NP>) (<L . POS POS . .>) ) ) ) 

In this example,
	- sentence: The finger-pointing has already begun .
	- chunks: (2, 3): has already (4, 4): begun
	- supertags: 
		The	NP[nb]/N
		finger-pointing	N
		has	(S[dcl]\NP)/(S[pt]\NP)
		already	(S\NP)\(S\NP)
		begun	S[pt]\NP
		.	.
	- CCG parsing result: (<T S[dcl] 1 2> (<T NP[nb] 0 2> (<L NP[nb]/N POS POS The NP[nb]/N>) (<L N POS POS finger-pointing N>) ) (<T S[dcl]\NP 0 2> (<T (S[dcl]\NP)/(S[pt]\NP) 0 2> (<L (S[dcl]\NP)/(S[pt]\NP) POS POS has (S[dcl]\NP)/(S[pt]\NP)>) (<L (S\NP)\(S\NP) POS POS already (S\NP)\(S\NP)>) ) (<T S[pt]\NP 0 2> (<L S[pt]\NP POS POS begun S[pt]\NP>) (<L . POS POS . .>) ) ) ) 

For more information, please read the appendix of our paper.
