I wasn't comparing that particular paper with malicious code. Inaccurate papers in other fields (like biology) might have a more significant impact.

If we were to continue the GitHub analogy, it would be great if ArXiv had a public "issue" list.