Malicious Action Detection on WebArena
7.46TPRGPT-OSS-Safeguard 20B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-OSS-Safeguard 20B2025.10 | 7.46 | 9.32 | |
| AprielGuard2025.10 | 3.19 | 0.62 | |
| Qwen-3-Guard2025.10 | 0.99 | 0.8 | |
| Granite-Guardian 3.3-8B2025.10 | 0.11 | 0.06 |
| Method | Links | ||
|---|---|---|---|
| GPT-OSS-Safeguard 20B2025.10 | 7.46 | 9.32 | |
| AprielGuard2025.10 | 3.19 | 0.62 | |
| Qwen-3-Guard2025.10 | 0.99 | 0.8 | |
| Granite-Guardian 3.3-8B2025.10 | 0.11 | 0.06 |