4 papers
Chain of Thought Monitoring Can Fail Even When CoT Is Necessary
Lennart Bürger, Piotr Komorowski, Moritz Weckbecker +4
Chain of Thought (CoT) monitoring is used by frontier labs to detect misbehaviour of AI systems. Yet, its reliability remains poorly understood. We attribute this to the lack of a…
Predicting Future Behaviors in Reasoning Models Enables Better Steering
Evgenii Kortukov, Piotr Komorowski, Florian Klein +5
Deployed large reasoning models (LRMs) often behave unexpectedly. Test-time steering controls LRM outputs by intervening on their hidden representations, but it can degrade output…
Strategic Dishonesty Can Undermine AI Safety Evaluations of Frontier LLMs
Alexander Panfilov, Evgenii Kortukov, Kristina Nikolić +6
Large language model (LLM) developers aim for their models to be honest, helpful, and harmless. However, when faced with malicious requests, models are trained to refuse, sacrifici…
ASIDE: Architectural Separation of Instructions and Data in Language Models
Egor Zverev, Evgenii Kortukov, Alexander Panfilov +5
Despite their remarkable performance, large language models lack elementary safety features, making them susceptible to numerous malicious attacks. In particular, previous work has…