English

Partial Answer of How Transformers Learn Automata

Formal Languages and Automata Theory 2025-05-13 v2 Machine Learning

Abstract

We introduce a novel framework for simulating finite automata using representation-theoretic semidirect products and Fourier modules, achieving more efficient Transformer-based implementations.

Keywords

Cite

@article{arxiv.2504.20395,
  title  = {Partial Answer of How Transformers Learn Automata},
  author = {Tiantian Zhang},
  journal= {arXiv preprint arXiv:2504.20395},
  year   = {2025}
}
R2 v1 2026-06-28T23:14:43.645Z