[Re] Pure Noise to the Rescue of Insufficient Data
cs.CV, cs.LG
Submitted: 2026-10-31
Updated: 2026-10-31
Code: https://github.com/seungjaeryanlee/pure-noise
Terminology
Sources
- Pure Noise to the Rescue of Insufficient Data: Improving Imbalanced Classification by Training on Random Noise Images
- Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss
- M2m: Imbalanced Classification via Major-to-minor Translation
- Wide Residual Networks
- A systematic study of the class imbalance problem in convolutional neural networks
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Class-Balanced Loss Based on Effective Number of Samples
- Improving Calibration for Long-Tailed Recognition
- Adversarial Examples Improve Image Recognition
- Improved Regularization of Convolutional Neural Networks with Cutout
- A Simple Framework for Contrastive Learning of Visual Representations
- AutoAugment: Learning Augmentation Policies from Data
- Generative Latent Implicit Conditional Optimization when Learning from Small Sample
- Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
- Deep Residual Learning for Image Recognition
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs