Tag: source code analysis

Researchers report that a transfer learning framework called TL-CDVD reached an average accuracy of 88.8 percent and an F1-score of ...

A new knowledge distillation method compresses large pre-trained models into lightweight detectors that identify vulnerabilities in Python code with Macro-F1 ...

A controlled benchmark of four pre-trained transformer models on the Devign dataset found CodeBERT achieved the highest mean accuracy, precision, ...

Researchers have developed a domain-adversarial deep learning method that captures contextual code semantics and aligns data distributions across projects, improving ...