All reviewers agree that this paper provides an original take on multi-attribute style transfer, with extensive experimentation. We believe this submission will be interesting to the NeurIPS audience and inspire fruitful research directions.