Universal and Transferable Adversarial Attacks on Aligned Language Models
Zou et al., 2023, Universal and Transferable Adversarial Attacks on Aligned Language Models (arXiv preprint arXiv:2307.15043)
Subscriber deck — free during early access
1 cards in “3 Experimental Results: Direct and Transfer Attacks” · whole deck
3 Experimental Results: Direct and Transfer Attacks
- What is AdvBench?Zou et al.’s benchmark with two settings: 500 harmful strings the model must output exactly (fine-grained control), and 500 harmful behaviors where any genuine attempt to comply counts as success (like red-teaming). The metric is attack success rate.
- Zou et al., 2023, Universal and Transferable Adversarial Attacks on Aligned Language Models (arXiv preprint arXiv:2307.15043), §3
Source
- Andy Zou, Zifan Wang, Nicholas Carlini, Milad Nasr, J. Zico Kolter and Matt Fredrikson. 2023. Universal and Transferable Adversarial Attacks on Aligned Language Models. In arXiv preprint arXiv:2307.15043. https://arxiv.org/abs/2307.15043