GICDM: Mitigating Hubness for Reliable Distance-Based Generative Model Evaluation
Nicolas Salvy, Hugues Talbot, Thirion Bertrand
Abstract
Generative model evaluation commonly relies on high-dimensional embedding spaces to compute distances between samples. We show that dataset representations in these spaces are affected by the hubness phenomenon, which distorts nearest neighbor relationships and biases distance-based metrics. Building on the classical Iterative Contextual Dissimilarity Measure (ICDM), we introduce Generative ICDM (GICDM), a method to correct neighborhood estimation for both real and generated data. We introduce a multi-scale extension to improve empirical behavior. Extensive experiments on synthetic and real benchmarks demonstrate that GICDM resolves hubness-induced failures, restores reliable metric behavior, and improves alignment with human judgment.
BibTeX
@inproceedings{
salvy2026gicdm,
title={{GICDM}: Mitigating Hubness for Reliable Distance-Based Generative Model Evaluation},
author={Nicolas Salvy and Hugues Talbot and Bertrand Thirion},
booktitle={Forty-third International Conference on Machine Learning},
year={2026},
url={https://openreview.net/forum?id=fJDZqUXTeE}
}