Open access
Aug 2026
Do Pre-Trained Code Models Add Value Beyond Software Metrics in Class-Level Defect Prediction? An Empirical Study of Input Coverage, Long-Code Aggregation, and Cross-Version Generalization
Findings indicate that pre-trained code models should be evaluated with explicit input-coverage reporting, chronological validation, strong metric baselines, and incremental-value testing.
Musaad Alzahrani
· Electronics · 0 citations