TinyImageNet

'TinyImageNet' is a dataset that contains a subset of 200 classes from the ImageNet dataset, each with 500 training images, used to evaluate the performance of machine learning models in visual recognition tasks.

Papers using TinyImageNet (2)

Do Transformers Need Three Projections? Systematic Study of QKV Variants2026Class Unlearning via Depth-Aware Removal of Forget-Specific Directions2026