Costa de Araujo, João PauloJoão PauloCosta de AraujoBalu, BalahariBalahariBaluReichmann, EikEikReichmannKelly, JessicaJessicaKellyKugele, StefanStefanKugeleMata, NúriaNúriaMataGrunske, LarsLarsGrunske2025-04-142025-04-142025https://publica.fraunhofer.de/handle/publica/48653810.18420/se2025-18In this extended abstract we summarize our work on using Concept Bottleneck Models (CBMs) for an enhanced safety argumentation of vision-based Machine Learning (ML) perception components in safety critical systems. This paper has been published at the International Symposium on Software Reliability Engineering (ISRRE’24).enconcept bottleneck modelsemantic gapsafetysafety assuranceinterpretabilityconference paper