Linear Representations of Hierarchical Concepts in Language Models

arXiv:2604.07886v1 Announce Type: new Abstract: We investigate how and to what extent hierarchical relations (e.g., Japan subset Eastern Asia subset Asia) are encoded in the internal representations of language models. Building on Linear Relational Concepts, we train linear transformations specific to each hierarchical depth and semantic domain, and characterize representational differences associated with hierarchical relations by comparing these transformations. Going beyond prior work on the representational geometry of hierarchies in LMs, our analysis covers multi-token entities and cross-layer representations. Across multiple domains we learn such transformations and evaluate in-domain generalization to unseen data and cross-domain transfer. Experiments show that, within a domain, hierarchical relations can be linearly recovered from model representations. We then analyze how hierarchical information is encoded in representation space. We find that it is encoded in a relatively low-dimensional subspace and that this subspace tends to be domain-specific. Our main result is that hierarchy representation is highly similar across these domain-specific subspaces. Overall, we find that all models considered in our experiments encode concept hierarchies in the form of highly interpretable linear representations.

Related

• What do Language Models Learn and When? The Implicit Curriculum Hypothesis
• Learning is Forgetting: LLM Training As Lossy Compression
• Emotion Concepts and their Function in a Large Language Model
• Current LLMs still cannot 'talk much' about grammar modules: Evidence from syntax

Source: arXiv cs.CL | 2026-04-10