Data span with syntactic parser

[youyouisme]

For the provided data in SNLI and ANLI, the sentence spans are already available. Have they been pre-processed as described in the DISCO paper using a neural syntactic parser? Could you provide the code for this preprocessing step?