StructLens: A Structural Lens for Language Models via Maximum Spanning Trees

ArXi:2603.03328v2 Announce Type: replace Language exhibits inherent structures, a property that explains both language acquisition and language change. Given this characteristic, we expect language models to manifest their own internal structures as well. While interpretability research has investigated how models compute representations mechanistically through attention patterns and Sparse AutoEncoders, the organization of the resulting representations is overlooked. To address this gap, we.