Backdoored PEFT adapters and limits of weight-only detection
Published Analyzed Source: arXiv
Third-party fine-tuning adapters may contain backdoors. Z-PEFT screens adapter weights using spectral features, evaluated on PADBench's 13,300 adapters.
Z-PEFT: Zero-shot Backdoor Detection in Parameter-Efficient Fine-Tuning via Canonical Spectral Signatures
Evaluated models: Llama 2 7B, Llama 2 13B, Qwen 1.5 7B+2 more
- Llama 2 7B
- Llama 2 13B
- Qwen 1.5 7B
- Flan-T5 XL
- RoBERTa Base