TIE: A Training-Inversion-Exclusion Framework for Visually Interpretable and Uncertainty-Guided Out-of-Distribution Detection

Explore the TIE framework, an advanced (n+1)-class architecture that leverages iterative inversion and exclusion to achieve interpretable, uncertainty-guided...

Level: advanced

By Pirzada Suhail, Rehna Afroz, Amit Sethi

Category: research