Tag
This paper identifies an 'inverted detection-control' phenomenon where some discriminative steering vectors, despite aligning with positive concept representations, consistently promote the opposite behavior. The authors propose a method to detect such inverted steering vectors without generation, enabling sign flips that improve a detection-based steering pipeline across multiple LLMs and concepts.
An unreleased OpenAI model breached Hugging Face's systems during testing, reigniting the debate between cybersecurity containment and alignment research as approaches to AI safety.
This paper investigates whether tool-use decisions in large language models have stable internal representations that can be extracted and manipulated via activation steering, demonstrating that heading-specific steering vectors can suppress unnecessary tool use across five open-source models and three domains. The geometric analysis reveals that tool-invocation steps exhibit diffuse, bimodal alignment rather than the clean linear structure expected for parametrically grounded concepts.