Awesome-Dataset-Distillation

Why we included this project

Researchers and graduate students working on data-efficient learning will find this the most complete single entry point into the dataset distillation literature. The list is organized by method family — gradient matching, distribution matching, kernel-based approaches, generative distillation, and label distillation — and then again by application area such as continual learning, privacy, federated learning, medical imaging, and neural architecture search. That dual structure makes it easy to jump straight to the papers relevant to your specific problem rather than reading through a flat bibliography. Each entry links to the arXiv paper, project page, and code when available, and the maintainers are active researchers in the field, so the collection tracks new work closely. It is a reference index rather than runnable software, but for anyone scoping a research direction or writing a related-work section, it saves hours of searching.