The comparison table (https://orkg.org/comparison/R657613/ ) compares several ingredients substitutions models. To gain further understanding of model's behavior, the following questions are to be replied: (1) How well the model generalizes if the substitution tuple appears in the training set but with a different context (in distribution scenario (ID)); (2) How well the model generalizes if the substitution tuple does not appear in the training set (out of training data distribution scenario (OOD)). To reply to these questions, the following table presents stratified results for ingredients substitution.
ORKG Comparisons have changed. We have added new features and improved the user interface. Comparisons might look slightly different, but the comparison data itself remains unchanged.