cs.CLNov 17, 2026

Structure--Residual Inductive Biases for Language Model Pretraining

Authors: Hai Huang, Bartosz Żuk, Rassul Magauin, Jiaxin Huang, Xue Liu

Organizations: Palona AI · Institute of Computer Science, Polish Academy of Sciences · Mohamed bin Zayed University of Artificial Intelligence

Abstract

Next-token prediction is relied upon to learn both predictive behavior and internal representations. We ask whether representation structure itself should be an explicit inductive bias. Motivated by the two-part coding view of Kolmogorov complexity, we study a structure--residual principle in which a representation separates structure from realization detail. We instantiate this principle in two complementary ways. Boundary Bottleneck Concept (BBC) routes information from token segments through boundary carriers while preserving local autoregressive computation. Continuous Concept (CC) regularizes representation geometry and induces a spectral decomposition into low-frequency structure and high-frequency residual. We derive coding bounds for both views and show that multiplicity-corrected boundary attention becomes exact as the within-segment path energy vanishes. Empirically, BBC and CC exhibit strikingly different depth preferences: BBC works best in early layers, whereas CC is most effective in later layers. At 300M training tokens, combining BBC and CC improves NLL by 1.44% on GPT-2 Medium. These gains persist with training runs up to 20B tokens. Representation diagnostics show that BBC increases the relative predictive importance of boundary states, while CC reduces spectral residual energy. Together, these results suggest that structure--residual biases can improve language-model pretraining beyond next-token prediction. Code: https://github.com/haihuang-ml/structure-residual

Figures & tables

Appendix figures & tables22 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

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

    Feb 10, 2026Haruki Sakajo, Frederikus Hudi, Yusuke Sakai +2LLM InterpretabilityRepresentation Geometry in Language Models

  2. Decoupling Token Roles in Autoregressive Pretraining

    Sep 27, 2026Suqin Yuan, Runqi Lin, Kevin Qinghong Lin +4Language Model PretrainingAutoregressive Language Modeling