Reviews: Maximizing Subset Accuracy with Recurrent Neural Networks in Multi-label Classification

I have been reviewing this paper for another conference, so in my review I mostly repeat my comments sent earlier. Already at that time I was for accepting the paper. It is worth to underline that the paper has been further improved by the authors since then. The paper considers a problem of solving multi-label classification (MLC) with recurrent neural networks (RNNs). MLC is converted into a sequential prediction problem: the model predicts a sequence of relevant labels.