Preprint
Aug 2026
SeFaR: Semantic Feature-aware Robustness Testing of Deep Neural Networks
Evaluation on case studies demonstrates that the proposed SeFaR framework effectively satisfies requirement preconditions while identifying requirement-independent features that influence model decisions, enabling it to both uncover faults and relate them to such features.
Nusrat Jahan Mozumder, Divya Gopinath, Corina S. Păsăreanu et al.
· 0 citations