PCA++: How Uniformity Induces Robustness to Background Noise in Contrastive Learning

NeurIPSSpotlight2025

Authors: Mingqi Wu, Qiang Sun, Archer Y. Yang
Venue: NeurIPS 2025
Track: Spotlight

TL;DR

High-dimensional data often conceal low-dimensional signals beneath structured background noise, limiting standard PCA…

Opening excerpt from the authors’ abstract. source

Read the paper

OpenReview

Topics

contrastive robustness

← All NeurIPS 2025 Spotlight papers · Browse the whole archive