English

The Diffusion-Attention Connection

Machine Learning 2026-04-14 v1

Abstract

Transformers, diffusion-maps, and magnetic Laplacians are usually treated as separate tools; we show they are all different regimes of a single Markov geometry built from pre-softmax query-scores. We define a QK "bidivergence" whose exponentiated and normalized forms yield attention, diffusion-maps, and magnetic diffusion. And use product of experts and Schr\"odinger-bridges to connect and organize them into equilibrium, nonequilibrium steady-state, and driven dynamics.

Cite

@article{arxiv.2604.09560,
  title  = {The Diffusion-Attention Connection},
  author = {Julio Candanedo},
  journal= {arXiv preprint arXiv:2604.09560},
  year   = {2026}
}
R2 v1 2026-07-01T12:03:17.417Z