| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| System Prompt Extraction | System Prompts 41-model | Self-Consistency Score0.915 | 10 | |
| Fingerprint Robustness Evaluation | System Prompts Weather | FSR100 | 9 | |
| Fingerprint Robustness Evaluation | System Prompts Pirate | FSR100 | 9 | |
| Fingerprint Robustness Evaluation | System Prompts OAI | FSR100 | 9 | |
| Fingerprint Robustness Evaluation | System Prompts Robot | FSR1 | 9 | |
| Multi-way Classification | 20 system prompts | SFT Accuracy63 | 6 | |
| Pairwise Classification | 20 system prompts | TF-IDF Accuracy97.1 | 6 |