Skip to content

Conversation

@mavroudisv
Copy link

This series of tasks measure the models' ability to detect (easy), classify (moderate) and patch (hard) software vulnerabilities in source code from 17 different programming languages.

Security vulnerabilities have been the subject of a very wide range of academic publications. However, the literature lacks a common point of reference making it hard to gauge progress in the field. We hope this task and dataset will serve as a common reference point motivating further steps forward.

All data fall under the Apache 2.0 license or other compatible licenses.

@mavroudisv
Copy link
Author

mavroudisv commented May 23, 2023

@Sohl-Dickstein any chance you could take a quick look in case there is anything we might have missed? :)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant