Data as of Aug 25, 2026 · Based on 3,181,687 AI responses across 10,525 prompts · See how Parse measures this
DATABenchgithub.comDATABench is an open-source benchmark framework for evaluating dataset auditing methods in deep learning from an adversarial perspective. It provides a unified evaluation pipeline and standardized interfaces to compare auditing algorithms and potential attacks, enabling fair and extensible research. The framework supports a three-stage attack pipeline (preprocessing, training, postprocessing) and ImageFolder-structured datasets, and includes a library of built-in attacks and auditing methods.
Sources
arxiv.org shapes more of what AI says about DATABench than any other source, at 83% of its citations.
chatpaper.com
The market map
AI Red Teaming Services →Where AI ranks DATABench
