Dear Editor, please find enclosed the manuscript, “Leveraging Neurosymbolic AI for Slice Discovery”, which we would like to submit for publication in the NeSy 2024 special issue of the Neurosymbolic Artificial Intelligence journal. In this paper, we propose a modular neurosymbolic AI approach for the slice discovery problem in Computer Vision (CV) models. Its distinctive advantage is the extraction via inductive logic programming of human-readable logical rules describing rare slices, and thus enhancing the explainability of CV models. To this end, we propose a methodology for inducing the occurrence of rare slices in a model. We present experiments conducted on datasets produced by our modified version of the Super-CLEVR data generator. The results show that our approach can correctly identify rare slices and produce logical rules describing them. The rules can be fruitfully used to generate new training data to mend model behaviour and thus enhance its inference capabilities. For these reasons, we think the paper could be of particular interest to the readers of the Neurosymbolic Artificial Intelligence journal. Looking forward to hearing from you, Michele Collevati, Thomas Eiter, Nelson Higuera Corresponding author: Michele Collevati Institute of Logic and Computation Technische Universität Wien Vienna, Austria Email: michele.collevati@tuwien.ac.at