PRISM: Recovering Instruction Sets from Language Model Activations 事件

PRISM: Recovering Instruction Sets from Language Model Activations arXiv:2606.09563v1 Announce Type: new Abstract: As LLMs are deployed as agents, reliable monitoring requires knowing not only what they output, but which instructions are steering their behavior. This is difficult when models infer unintended subgoals, follow contextual cues, or are influenced by prompt injections and hidden objectives. While activation-to-language methods suggest that hidden states can reveal natural-language i