GitHub’s model for secret detection: Finding more secrets with less noise
The GitHub Blog published the article ‘GitHub’s model for secret detection: Finding more secrets with less noise’. The model uses a fine‑tuned ModernBERT classifier for context‑aware secret classification. The article describes GitHub secret scanning and its push‑protection and post‑push‑scanning modes. The model improves secret scanning by increasing weighted acceptance by 11.5 % and reducing median scan execution time by 86 %.· Entity summary generated by Graphiti
Relation timeline
BY VALID-FROM DATE- GitHub’s model for secret detection: Finding more secrets with less noiseUSESModernBERT classifier
The model uses a fine‑tuned ModernBERT classifier for context‑aware secret classification.
- GitHub’s model for secret detection: Finding more secrets with less noiseDESCRIBESGitHub Secret Protection
The article describes GitHub secret scanning and its push‑protection and post‑push‑scanning modes.
- GitHub’s model for secret detection: Finding more secrets with less noiseIMPROVEDGitHub Secret Protection
The model improves secret scanning by increasing weighted acceptance by 11.5 % and reducing median scan execution time by 86 %.
- GitHub BlogPUBLISHED ARTICLEGitHub’s model for secret detection: Finding more secrets with less noise
The GitHub Blog published the article ‘GitHub’s model for secret detection: Finding more secrets with less noise’.
Relations are facts Graphiti extracted from reports, not verbatim quotes. An end date means later reporting showed the relation no longer holds.