Phylogenetic Tree Inference with Tropical Axial Attention

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Teska, Chris, Pasque, Kurt, Yoshida, Ruriko, Hashemi, Baran
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918500638916608
author Teska, Chris
Pasque, Kurt
Yoshida, Ruriko
Hashemi, Baran
author_facet Teska, Chris
Pasque, Kurt
Yoshida, Ruriko
Hashemi, Baran
contents In this work, we introduce a Tropical Axial Attention neural reasoning architecture that replaces vanilla softmax dot-product attention with max-plus operators, inducing a piecewise-linear structure aligned with dynamic programming formulations. From multi-species sequence alignments, our model learns all possible pairwise distances and is trained using a combination of $\ell_1$ and tropical symmetric distance metric losses with an ultrametric violation penalty. We leverage the well known isomorphic relationship between the space of all phylogenetic trees with $n$ species and tropical Grassmannian to show that tropical attention provides a natural geometric framework for phylogenetic inference. On empirical $DS1-DS11$ alignments, where true trees are unknown, the tropical model produces distance matrices that are substantially closer to their BME-induced tree metrics than the baseline models. These results suggest that tropical attention is a useful geometric inductive bias for neural phylogenetic inference, especially under distribution shift and when tree-metric consistency is important.
format Preprint
id arxiv_https___arxiv_org_abs_2605_13894
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Phylogenetic Tree Inference with Tropical Axial Attention
Teska, Chris
Pasque, Kurt
Yoshida, Ruriko
Hashemi, Baran
Populations and Evolution
Machine Learning
In this work, we introduce a Tropical Axial Attention neural reasoning architecture that replaces vanilla softmax dot-product attention with max-plus operators, inducing a piecewise-linear structure aligned with dynamic programming formulations. From multi-species sequence alignments, our model learns all possible pairwise distances and is trained using a combination of $\ell_1$ and tropical symmetric distance metric losses with an ultrametric violation penalty. We leverage the well known isomorphic relationship between the space of all phylogenetic trees with $n$ species and tropical Grassmannian to show that tropical attention provides a natural geometric framework for phylogenetic inference. On empirical $DS1-DS11$ alignments, where true trees are unknown, the tropical model produces distance matrices that are substantially closer to their BME-induced tree metrics than the baseline models. These results suggest that tropical attention is a useful geometric inductive bias for neural phylogenetic inference, especially under distribution shift and when tree-metric consistency is important.
title Phylogenetic Tree Inference with Tropical Axial Attention
topic Populations and Evolution
Machine Learning
url https://arxiv.org/abs/2605.13894