Anomalicious: Automated Detection of Anomalous and Potentially Malicious Commits on GitHub · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2103.03846/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTAnomalicious: Automated Detection of Anomalous and Potentially Malicious Commits on GitHub — mentioned by kekebohttps://news.ycombinator.com/item?id=30790798urn:hackerpapers:mention:2103.03846:30790798Thu, 24 Mar 2022 14:58:45 GMT<p>Could (/should) this be mitigated on the repository host side by scanning for and flagging malicious commits? A paper from last year evaluating this on Github achieves a ~50% success rate[0]. Given Github already training ml models across all repositories for Copilot I would guess higher rates would be possible. [0] https://arxiv.org/abs/2103.03846 Edit: add link</p><p><a href="https://hackerpaper.kaipfister.de/p/2103.03846/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.SE