Copertina

Understanding the Detrimental Class-level Effects of Data Augmentation

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Kirichenko, Polina, Ibrahim, Mark, Balestriero, Randall, Bouchacourt, Diane, Vedantam, Ramakrishna, Firooz, Hamed, Wilson, Andrew Gordon
Natura:	Preprint
Pubblicazione:	2023
Soggetti:	Computer Vision and Pattern Recognition Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

UniBench: Visual Reasoning Requires Rethinking Vision-Language Beyond Scaling
di: Al-Tahan, Haider, et al.
Pubblicazione: (2024)

$\mathbb{X}$-Sample Contrastive Loss: Improving Contrastive Learning with Sample Similarity Graphs
di: Sobal, Vlad, et al.
Pubblicazione: (2024)

Self-Supervised Disentanglement by Leveraging Structure in Data Augmentations
di: Eastwood, Cian, et al.
Pubblicazione: (2023)

Embracing Diversity: Interpretable Zero-shot classification beyond one vector per class
di: Moayeri, Mazda, et al.
Pubblicazione: (2024)

Modeling Caption Diversity in Contrastive Vision-Language Pretraining
di: Lavoie, Samuel, et al.
Pubblicazione: (2024)

What's in Common? Multimodal Models Hallucinate When Reasoning Across Scenes
di: Ross, Candace, et al.
Pubblicazione: (2025)

VISReg: Variance-Invariance-Sketching Regularization for JEPA training
di: Wu, Haiyu, et al.
Pubblicazione: (2026)

Hyperbolic Image-Text Representations
di: Desai, Karan, et al.
Pubblicazione: (2023)

Post-Hoc Guidance for Consistency Models by Joint Flow Distribution Learning
di: Hsu, Chia-Hong, et al.
Pubblicazione: (2026)

Joint Embedding vs Reconstruction: Provable Benefits of Latent Space Prediction for Self Supervised Learning
di: Van Assel, Hugues, et al.
Pubblicazione: (2025)

Retrievals Can Be Detrimental: Unveiling the Backdoor Vulnerability of Retrieval-Augmented Diffusion Models
di: Fang, Hao, et al.
Pubblicazione: (2025)

LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristics
di: Balestriero, Randall, et al.
Pubblicazione: (2025)

Learning by Reconstruction Produces Uninformative Features For Perception
di: Balestriero, Randall, et al.
Pubblicazione: (2024)

Beyond and Free from Diffusion: Invertible Guided Consistency Training
di: Hsu, Chia-Hong, et al.
Pubblicazione: (2025)

From Linearity to Non-Linearity: How Masked Autoencoders Capture Spatial Correlations
di: Bisulco, Anthony, et al.
Pubblicazione: (2025)

Self-supervised Video Instance Segmentation Can Boost Geographic Entity Alignment in Historical Maps
di: Xia, Xue, et al.
Pubblicazione: (2024)

ViDDAR: Vision Language Model-Based Task-Detrimental Content Detection for Augmented Reality
di: Xiu, Yanming, et al.
Pubblicazione: (2025)

Self-Supervised Anomaly Detection in the Wild: Favor Joint Embeddings Methods
di: Otero, Daniel, et al.
Pubblicazione: (2024)

Visual Compositional Tuning
di: Wu, Xindi, et al.
Pubblicazione: (2025)

Gaussian Embeddings: How JEPAs Secretly Learn Your Data Density
di: Balestriero, Randall, et al.
Pubblicazione: (2025)

Ditch the Denoiser: Emergence of Noise Robustness in Self-Supervised Learning from Data Curriculum
di: Lu, Wenquan, et al.
Pubblicazione: (2025)

MapSAM2: Adapting SAM2 for Automatic Segmentation of Historical Map Images and Time Series
di: Xia, Xue, et al.
Pubblicazione: (2025)

Deep Networks Always Grok and Here is Why
di: Humayun, Ahmed Imtiaz, et al.
Pubblicazione: (2024)

On the Geometry of Deep Learning
di: Balestriero, Randall, et al.
Pubblicazione: (2024)

The Impact of Coreset Selection on Spurious Correlations and Group Robustness
di: Dharmasiri, Amaya, et al.
Pubblicazione: (2025)

SplineCam: Exact Visualization and Characterization of Deep Network Geometry and Decision Boundaries
di: Humayun, Ahmed Imtiaz, et al.
Pubblicazione: (2023)

The Effects of Mixed Sample Data Augmentation are Class Dependent
di: Lee, Haeil, et al.
Pubblicazione: (2023)

Beyond [cls]: Exploring the true potential of Masked Image Modeling representations
di: Przewięźlikowski, Marcin, et al.
Pubblicazione: (2024)

DIET-CP: Lightweight and Data Efficient Self Supervised Continued Pretraining
di: Rodas, Bryan, et al.
Pubblicazione: (2025)

GPS-SSL: Guided Positive Sampling to Inject Prior Into Self-Supervised Learning
di: Feizi, Aarash, et al.
Pubblicazione: (2024)

Class-specific Data Augmentation for Plant Stress Classification
di: Saleem, Nasla, et al.
Pubblicazione: (2024)

WILD SAM: A Simulated-and-Real Data Augmentation for Autonomous Driving Perception under Challenging Weather
di: Khatounabadi, Hamed, et al.
Pubblicazione: (2026)

Emphasizing Discriminative Features for Dataset Distillation in Complex Scenarios
di: Wang, Kai, et al.
Pubblicazione: (2024)

FastDINOv2: Frequency Based Curriculum Learning Improves Robustness and Training Speed
di: Zhang, Jiaqi, et al.
Pubblicazione: (2025)

Rectified LpJEPA: Joint-Embedding Predictive Architectures with Sparse and Maximum-Entropy Representations
di: Kuang, Yilun, et al.
Pubblicazione: (2026)

Understanding Open-Set Recognition by Jacobian Norm and Inter-Class Separation
di: Park, Jaewoo, et al.
Pubblicazione: (2022)

Walking the Tightrope: Disentangling Beneficial and Detrimental Drifts in Non-Stationary Custom-Tuning
di: Yang, Xiaoyu, et al.
Pubblicazione: (2025)

MADrive: Memory-Augmented Driving Scene Modeling
di: Karpikova, Polina, et al.
Pubblicazione: (2025)

RA-Touch: Retrieval-Augmented Touch Understanding with Enriched Visual Data
di: Cho, Yoorhim, et al.
Pubblicazione: (2025)

Rethinking Video-Text Understanding: Retrieval from Counterfactually Augmented Data
di: Ma, Wufei, et al.
Pubblicazione: (2024)