The journal paper “Learning Program Behavioral Models from Synthesized Input-Output Pairs” was accepted at the ACM Transactions on Software Engineering and Methodology (TOSEM) today. In this paper, Tural Mammadov at CISPA shows how to systematically generate pairs of inputs and outputs for complex programs and train a neural model on them. The model achieves high accuracy on predicting outputs from inputs, but in particular also can predict inputs from outputs, inverting the behavior of the original program. Congratulations, Tural!