Particularly for judged tasks, but also for others, it certainly would be beneficial if the model not only includes targets and hints, but also answers. The difference between an answer and a target is, that the answer could be displayed along the judgement view and when the task is over.
May we take this feature request and consider partial ground-truth tasks, which currently are not really supported.