Emergence and Localisation of Semantic Role Circuits in LLMs

ACL ARR 2026 January Submission5705 Authors

Keywords: semantic roles, circuit discovery, mechanistic interpretability, large language models, training dynamics, causal analysis

Abstract: Despite displaying semantic competence, large language models' internal mechanisms that ground abstract semantic structure remain insufficiently characterised. To investigate whether and how LLMs develop causally functional representations of semantic roles, we introduce a causal-temporal methodology combining contrastive minimal pairs, edge-attribution circuit discovery, and training-time tracking. Our analysis reveals that LLMs encode semantic roles through highly localised circuits (89–92\% attribution within $\leq$28 nodes) that emerge gradually via structural refinement rather than phase transitions. These circuits exhibit moderate cross-scale conservation (24–51\% component overlap) alongside high spectral similarity, with larger models reusing similar components while rewiring connections. These findings suggest that LLMs form compact, causally isolated mechanisms for abstract semantic structure that exhibit partial transfer across scales and architectures.

Paper Type: Long

Research Area: Interpretability and Analysis of Models for NLP

Research Area Keywords: Interpretability and Analysis of Models for NLP

Contribution Types: Model analysis & interpretability

Languages Studied: English

Submission Number: 5705