Feature Whitening

Follow this topic Ask about this topic

Latest papers

10 papers2 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Whitening Improves Robustness to Spurious Correlations in Linear Probes
Sep 30, 2026Floris Holstege, Bram Wouters, Noud van Giersbergen +1Linear Probing Feature Whitening
University of Amsterdam, Department of Quantitative Economics · Tinbergen Institute
WhiteCon: Semi-Supervised Domain Adaptation Regression Through Whitening Transform and Dual Consistency
Sep 28, 2026Se Jin Sim, Seoung Bum KimDomain Adaptation Feature Whitening
School of Industrial and Management Engineering Korea University, Seoul, Republic of Korea
Whitening Inverts the Hierarchy: What the Norm of a Whitened Embedding Measures
Sep 19, 2026Mohammed Ahnouch, Lotfi ElaachakFeature Whitening OOD Detection
Université Paris 1 Paris, France · Faculty of Science and Technology of Tangier Abdelmalek Essaadi University Tangier, Morocco
Calibrating WEAT Against Anisotropy: ZCA Whitening as a Geometric Pre-Processing Step for Embedding Association Tests
Aug 7, 2026Seitaro Ono, Senna Ross, Jun SaikiNeural Representation Geometry Text Embedding Evaluation
Graduate School of Human and Environmental Studies, Kyoto University, Kyoto, Japan · Brock University, St. Catharines, Ontario, Canada
A Theory of Contrastive Learning with Natural Images
Jul 8, 2026Antonio Torralba, Yair WeissContrastive Learning Frequency-Domain Feature Learning
CSAIL, MIT · School of Computer Science and Engineering, Hebrew University of Jerusalem
Zeta: Dual Whitening for Matrix Optimization via Coordinate-Adaptive Preconditioning
Jun 12, 2026Kaiwen Chen, Shuhai Zhang, Zimo Liu +7Neural Network Optimization Feature Whitening
South China University of Technology · AIGCode · Hong Kong Baptist University
Improving clinical interpretability of linear neuroimaging models through feature whitening
Apr 22, 2026Sara Petiton, Antoine Grigis, Raphaël Vock +1Medical Diagnosis Feature Whitening
University Paris-Saclay, CEA, CNRS, NeuroSpin, Baobab UMR 9027, Saclay, France
Voronoi-based Second-order Descriptor with Whitened Metric in LiDAR Place Recognition
Mar 16, 2026Jaein Kim, Hee Bin Yoo, Dong-Sig Han +1Feature Whitening LiDAR Place Recognition
Interdisciplinary Program in Neuroscience, Seoul National University · Département d’Informatique, École Normale Supérieure (ENS) · Department of Computing, Imperial College London +1
Towards Generalizable Deepfake Detection via Real Distribution Bias Correction
Mar 14, 2026Ming-Hui Liu, Harry Cheng, Xin Luo +2Feature Whitening Deepfake Detection
School of Software, Shandong University, Jinan 250101, China · School of Computing, National University of Singapore, Singapore · School of Software, Shandong University, Jinan 250100, China
Whitening Reveals Cluster Commitment as the Geometric Separator of Hallucination Types
Mar 8, 2026Matic KorunFeature Whitening Hallucination in Language Models
Independent Researcher, Ljubljana, Slovenia

Sep 30, 2026J/K move · Enter open · S save

Whitening Improves Robustness to Spurious Correlations in Linear Probes

Floris Holstege, Bram Wouters, Noud van Giersbergen, Cees Diks

University of Amsterdam, Department of Quantitative Economics · Tinbergen Institute

Linear Probing Feature Whitening Spurious Correlation Robustness Spurious Correlation

Open paper AskPDF arXiv

Deep neural networks tend to rely on simple features that may be spurious and thus fail to generalize. We study this problem in the setting of linear probes, where a (generalized) linear model is fitted on the representations of a (pretrained) model. We use the connection of these models to the max-margin classifier, and show they favor directions associated with large eigenvalues of the covariance matrix. Whitening removes this preference by equalizing the eigenvalues of the covariance matrix. This observation motivates whitening as a preprocessing step that can reduce reliance on spurious correlations without requiring prior knowledge of their presence or labeled data. We examine the effect of whitening on a synthetic data-generating process and standard spurious correlation benchmarks, and find that it improves robustness. We also find that whitening can improve robustness when added to existing approaches.