H-Probes: Extracting Hierarchical Structures From Latent Representations of Language Models

arXiv:2605.00847v1 Announce Type: new Abstract: Representing and navigating hierarchy is a fundamental primitive of reasoning. Large language models have demonstrated proficiency in a wide variety of tasks requiring hierarchical reasoning, but there exists limited analysis on how the models geometrically represent the necessary latent constructions for such thinking. To this end, we develop \textit{H-probes}, a collection of linear probes that extract hierarchical structure, specifically depth a...