Backdoor Attack on CodeAlpaca-20k (test)
90.5ASRBadDLM
Evaluation Results
| Method | Links | |
|---|---|---|
| BadDLMMethod=BadDLM (Ours), Base Model=Dream-Instruct-7B2026.05 | 90.5 | |
| RL-basedMethod=RL-based, Base Model=Dream-Instruct-7B2026.05 | 70.9 | |
| SFT-basedMethod=SFT-based, Base Model=Dream-Instruct-7B2026.05 | 53.8 | |
| VPIMethod=VPI, Base Model=Dream-Instruct-7B2026.05 | 47.4 | |
| Benign (No Attack)Method=Benign (No Attack), Base Model=Dream-Instruct-7B2026.05 | 0 |