TailBooster Framework Enhances Extreme Value Augmentation in Air Transport
August 11, 2026
TailBooster is a dual-layer generative framework designed to augment rare, extreme events in air transport data using generative modeling and anomaly detection. It ensures operational validity, preventing infeasible synthetic instances like unrealistic flight times.
HOW THIS AFFECTS YOU
●
researcherUse this framework to address distributional tail issues in mixed-type tabular datasets.
●
healthSimilar dual-layer approaches could be applied to rare clinical event modeling in healthcare.