MechELK: A Mechanistic Interpretability Framework for Eliciting Latent Knowledge in Large Language Models 事件

MechELK: A Mechanistic Interpretability Framework for Eliciting Latent Knowledge in Large Language Models arXiv:2605.28825v1 Announce Type: new Abstract: Large language models (LLMs) frequently encode factual and reasoning knowledge in their internal representations that is not faithfully reflected in their surface-level outputs -- a phenomenon known as \emph{latent knowledge}. Existing approaches to eliciting latent knowledge, such as Contrastive Consistency Search (CCS), rely on contrastive a