Identification of OWASP Input Validation secure coding practices on WSVD-Bench
94.94Accuracygpt-4.1
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| gpt-4.1Prompting strategy=zero-shot, Context=Prompt 1 (Baseline)2026.03 | 94.94 | 97.9 | 85.09 | 89.32 | |
| gpt-4.1-miniPrompting strategy=zero-shot, Context=Prompt 1 (Baseline)2026.03 | 88.24 | 84.93 | 82.9 | 80.8 | |
| gemini-2.5-flashPrompting strategy=zero-shot, Context=Prompt 1 (Baseline)2026.03 | 84.52 | 81.46 | 80.01 | 77.32 | |
| gpt-4o-miniPrompting strategy=zero-shot, Context=Prompt 1 (Baseline)2026.03 | 65.48 | 73.8 | 69.61 | 60.92 | |
| gpt-3.5-turboPrompting strategy=zero-shot, Context=Prompt 1 (Baseline)2026.03 | 54.76 | 54.13 | 71.21 | 52.13 |