Function-level vulnerability detection on PrimeVul Paired
53.22AccuracyGLM-4.7
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| GLM-4.7decoding=greedy2026.02 | 53.22 | 53.24 | 58.62 | 55.8 | 20.55 | 38.6 | 26.57 | 12.53 | |
| Claude-Opus-4-5decoding=greedy2026.02 | 52.73 | 52.42 | 59.02 | 55.53 | 9.02 | 50.06 | 37.28 | 3.63 | |
| Claude-sonnet-4-5decoding=greedy2026.02 | 52.57 | 51.44 | 91.8 | 65.93 | 8.52 | 83.38 | 4.93 | 3.17 | |
| Deepseek-V3.2decoding=greedy2026.02 | 50.09 | 51.27 | 4.89 | 8.91 | 3.51 | 1 | 91.6 | 3.88 | |
| Qwen3-Coder-Nextdecoding=greedy2026.02 | 48.33 | 48.54 | 54.54 | 51.37 | 0.88 | 53.01 | 41.29 | 4.64 |