Randall Balestriero

papers

492

total citations

papers (20)

Contrastive and Non-Contrastive Self-Supervised Learning Recover Global and Local Spectral Embedding Methods

NEURIPS 2022arXiv

158

citations

Polarity Sampling: Quality and Diversity Control of Pre-Trained Generative Networks via Singular Values

CVPR 2022arXiv

citations

SplineCam: Exact Visualization and Characterization of Deep Network Geometry and Decision Boundaries

CVPR 2023arXiv

citations

Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models

NEURIPS 2025arXiv

citations

Understanding the detrimental class-level effects of data augmentation

NEURIPS 2023arXiv

citations

Active Self-Supervised Learning: A Few Low-Cost Relationships Are All You Need

ICCV 2023arXiv

citations

$\mathbb{X}$-Sample Contrastive Loss: Improving Contrastive Learning with Sample Similarity Graphs

ICLR 2025arXiv

citations

Characterizing Large Language Model Geometry Helps Solve Toxicity Detection and Generation

ICML 2024arXiv

citations

Beyond [cls]: Exploring the True Potential of Masked Image Modeling Representations

ICCV 2025arXiv

citations

Curvature Tuning: Provable Training-free Model Steering From a Single Parameter

NEURIPS 2025arXiv

citations

FastDINOv2: Frequency Based Curriculum Learning Improves Robustness and Training Speed

NEURIPS 2025arXiv

citations

Ditch the Denoiser: Emergence of Noise Robustness in Self-Supervised Learning from Data Curriculum

NEURIPS 2025arXiv

citations

A Data-Augmentation Is Worth A Thousand Samples: Analytical Moments And Sampling-Free Training

NEURIPS 2022

citations

Analytical Probability Distributions and Exact Expectation-Maximization for Deep Generative Networks

NEURIPS 2020

citations

From Linearity to Non-Linearity: How Masked Autoencoders Capture Spatial Correlations

ICCV 2025arXiv

citations

How Learning by Reconstruction Produces Uninformative Features For Perception

ICML 2024

citations

An Information Theory Perspective on Variance-Invariance-Covariance Regularization

NEURIPS 2023

citations

Randall Balestriero

papers (20)

Contrastive and Non-Contrastive Self-Supervised Learning Recover Global and Local Spectral Embedding Methods

The Effects of Regularization and Data Augmentation are Class Dependent

Deep Networks Always Grok and Here is Why

projUNN: efficient method for training deep networks with unitary matrices

Polarity Sampling: Quality and Diversity Control of Pre-Trained Generative Networks via Singular Values

SplineCam: Exact Visualization and Characterization of Deep Network Geometry and Decision Boundaries

Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models

Understanding the detrimental class-level effects of data augmentation

Active Self-Supervised Learning: A Few Low-Cost Relationships Are All You Need

$\mathbb{X}$-Sample Contrastive Loss: Improving Contrastive Learning with Sample Similarity Graphs

Characterizing Large Language Model Geometry Helps Solve Toxicity Detection and Generation

Beyond [cls]: Exploring the True Potential of Masked Image Modeling Representations

Curvature Tuning: Provable Training-free Model Steering From a Single Parameter

FastDINOv2: Frequency Based Curriculum Learning Improves Robustness and Training Speed

Ditch the Denoiser: Emergence of Noise Robustness in Self-Supervised Learning from Data Curriculum

A Data-Augmentation Is Worth A Thousand Samples: Analytical Moments And Sampling-Free Training

Analytical Probability Distributions and Exact Expectation-Maximization for Deep Generative Networks

From Linearity to Non-Linearity: How Masked Autoencoders Capture Spatial Correlations

How Learning by Reconstruction Produces Uninformative Features For Perception

An Information Theory Perspective on Variance-Invariance-Covariance Regularization

papers (20)

Contrastive and Non-Contrastive Self-Supervised Learning Recover Global and Local Spectral Embedding Methods

The Effects of Regularization and Data Augmentation are Class Dependent

Deep Networks Always Grok and Here is Why

projUNN: efficient method for training deep networks with unitary matrices

Polarity Sampling: Quality and Diversity Control of Pre-Trained Generative Networks via Singular Values

SplineCam: Exact Visualization and Characterization of Deep Network Geometry and Decision Boundaries

Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models

Understanding the detrimental class-level effects of data augmentation

Active Self-Supervised Learning: A Few Low-Cost Relationships Are All You Need

$\mathbb{X}$-Sample Contrastive Loss: Improving Contrastive Learning with Sample Similarity Graphs

Characterizing Large Language Model Geometry Helps Solve Toxicity Detection and Generation

Beyond [cls]: Exploring the True Potential of Masked Image Modeling Representations

Curvature Tuning: Provable Training-free Model Steering From a Single Parameter

FastDINOv2: Frequency Based Curriculum Learning Improves Robustness and Training Speed

Ditch the Denoiser: Emergence of Noise Robustness in Self-Supervised Learning from Data Curriculum

A Data-Augmentation Is Worth A Thousand Samples: Analytical Moments And Sampling-Free Training

Analytical Probability Distributions and Exact Expectation-Maximization for Deep Generative Networks

From Linearity to Non-Linearity: How Masked Autoencoders Capture Spatial Correlations

How Learning by Reconstruction Produces Uninformative Features For Perception

An Information Theory Perspective on Variance-Invariance-Covariance Regularization