Portrait of Linh Le is unavailable

Linh Le

Independent visiting researcher - University of Technology Sydney
Supervisor
Adam M. Oberman
Research Topics
AI Safety
Google Scholar
GitHub

Publications

Can Safety Fine-Tuning Be More Principled? Lessons Learned from Cybersecurity
David Williams-King
Linh Le
Adam Oberman
Yoshua Bengio
As LLMs develop increasingly advanced capabilities, there is an increased need to minimize the harm that could be caused to society by certa… (see more)in model outputs; hence, most LLMs have safety guardrails added, for example via fine-tuning. In this paper, we argue the position that current safety fine-tuning is very similar to a traditional cat-and-mouse game (or arms race) between attackers and defenders in cybersecurity. Model jailbreaks and attacks are patched with bandaids to target the specific attack mechanism, but many similar attack vectors might remain. When defenders are not proactively coming up with principled mechanisms, it becomes very easy for attackers to sidestep any new defenses. We show how current defenses are insufficient to prevent new adversarial jailbreak attacks, reward hacking, and loss of control problems. In order to learn from past mistakes in cybersecurity, we draw analogies with historical examples and develop lessons learned that can be applied to LLM safety. These arguments support the need for new and more principled approaches to designing safe models, which are architected for security from the beginning. We describe several such approaches from the AI literature.
2025-01-19
ArXiv (preprint)
arxiv.org
arxiv.org
Can Safety Fine-Tuning Be More Principled? Lessons Learned from Cybersecurity
David Williams-King
Linh Le
Adam M. Oberman
Yoshua Bengio
As LLMs develop increasingly advanced capabilities, there is an increased need to minimize the harm that could be caused to society by certa… (see more)in model outputs; hence, most LLMs have safety guardrails added, for example via fine-tuning. In this paper, we argue the position that current safety fine-tuning is very similar to a traditional cat-and-mouse game (or arms race) between attackers and defenders in cybersecurity. Model jailbreaks and attacks are patched with bandaids to target the specific attack mechanism, but many similar attack vectors might remain. When defenders are not proactively coming up with principled mechanisms, it becomes very easy for attackers to sidestep any new defenses. We show how current defenses are insufficient to prevent new adversarial jailbreak attacks, reward hacking, and loss of control problems. In order to learn from past mistakes in cybersecurity, we draw analogies with historical examples and develop lessons learned that can be applied to LLM safety. These arguments support the need for new and more principled approaches to designing safe models, which are architected for security from the beginning. We describe several such approaches from the AI literature.
2024-10-12
NeurIPS.cc/2024/Workshop/SafeGenAi (poster)
doi.org
openreview.net
Can Safety Fine-Tuning Be More Principled? Lessons Learned from Cybersecurity
David Williams-King
Linh Le
Adam M. Oberman
Yoshua Bengio
As LLMs develop increasingly advanced capabilities, there is an increased need to minimize the harm that could be caused to society by certa… (see more)in model outputs; hence, most LLMs have safety guardrails added, for example via fine-tuning. In this paper, we argue the position that current safety fine-tuning is very similar to a traditional cat-and-mouse game (or arms race) between attackers and defenders in cybersecurity. Model jailbreaks and attacks are patched with bandaids to target the specific attack mechanism, but many similar attack vectors might remain. When defenders are not proactively coming up with principled mechanisms, it becomes very easy for attackers to sidestep any new defenses. We show how current defenses are insufficient to prevent new adversarial jailbreak attacks, reward hacking, and loss of control problems. In order to learn from past mistakes in cybersecurity, we draw analogies with historical examples and develop lessons learned that can be applied to LLM safety. These arguments support the need for new and more principled approaches to designing safe models, which are architected for security from the beginning. We describe several such approaches from the AI literature.
2024-10-12
NeurIPS.cc/2024/Workshop/SafeGenAi (poster)
openreview.net
openreview.net