could you provide the whole original datasets？

[ruanruan0518]

could you provide the whole original datasets？

[ruanruan0518]

and provide the data precessing script？

[br0kej]

Hey!

The original dataset used for training is Dataset-1 from Cisco Talos's Binary Similarity paper. The repo is here: https://github.com/Cisco-Talos/binary_function_similarity

The data processing used Bin2ml which is also on GitHub. It can be found here: https://github.com/br0kej/bin2ml

To generate the data, you use bin2ml to extract the cfg data type -m cfg and then use bin2ml to generate esil function strings.

You should then end up with a processed dataset that looks like the one linked in the Section 4 of the README.

Hope that helps!