cs.LGNov 5, 2026

MEDALLION: Manifold Embedding Distillation and Attributions for Low-dimensional Latent InterpretatiON

Authors: Vicky Jung, Claire M. He, Irene Chang, Genevera I. Allen

Organizations: University of Florida, Gainsville, USA · Department of Statistics, Columbia University, New York, USA

Abstract

Dimensionality reduction is widely used to visualize high-dimensional data and guide scientific discovery, yet interpreting which original features contribute to low-dimensional representations remains challenging, particularly for black-box methods such as t-SNE and UMAP. We introduce MEDALLION (Manifold Embedding Distillation and Attributions for Low-dimensional Latent InterpretatiON), a novel framework for interpreting low-dimensional embeddings through knowledge distillation and neural feature attribution. MEDALLION distills any given embedding with a neural encoder which we explain via a broad range of existing neural feature attribution methods. Through experiments on MNIST and applications in single-cell transcriptomics, we show MEDALLION provides informative feature attributions that cast additional insights into the underlying embedding and compare its performance with existing interpretability approaches.

Figures & tables

Explore similar work

CardsList
  1. MEDAL: Manifold Embedding Distillation via Autoencoder Learning

    May 22, 2026Irene Chang, Tarek M. Zikry, Genevera I. AllenAutoencodersManifold Learning

  2. FADEx: Feature Attribution and Distortion-based Explanation of Dimensionality Reduction

    Jul 29, 2026Lucas Greff Meneses, Evandro S. Ortigossa, Claudio Silva +1Feature AttributionInverse Dimensionality Reduction

  3. The Rashomon Effect for Visualizing High-Dimensional Data

    Apr 1, 2026Yiyang Sun, Haiyang Huang, Gaurav Rajesh Parikh +1Dimensionality Reduction