Forest-Guided Semantic Transport for Label-Supervised Manifold Alignment

Adrien Aumon, Myriam Lizotte, Guy Wolf, Kevin R. Moon, Jake S. Rhodes · 2026

Read Forest-Guided Semantic Transport for Label-Supervised Manifold Alignment by Adrien Aumon, Myriam Lizotte, Guy Wolf, Kevin R. Moon, Jake S. Rhodes (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-00974.txt.

Description

Adrien Aumon and colleagues align data from varied sources by learning label shapes inside each source first. Their FoSTA method builds supervised forest geometries per domain, then uses shared class meaning to link domains without preset anchors. It extends RF-GAP affinities to partly labeled data and lines up the resulting semantic maps through scalable hierarchical transport. Tests against known baselines show strong recovery of true matches and preserved semantic structure. The method proves useful on synthetic and real sets, including multimodal integration and single-cell batch correction.

Catalog record

CREATORAdrien Aumon, Myriam Lizotte, Guy Wolf, Kevin R. Moon, Jake S. Rhodes
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE54911
LICENSEhttp://creativecommons.org/licenses/by-sa/4.0/
RIGHTSCC BY SA
SOURCEarXiv

Source: https://arxiv.org/abs/2602.00974

Browse the full library · About the Tower of Babel