Possibly, but I didn't see anything in the paper about energy efficiency vs a similar software implementation of their neural net and I don't think the wider Deep Learning community really has spent much time thinking about energy efficiency of their models.
There is a substantial subfield doing embedded deep learning - they absolutely care about energy efficiency.
"Substantial", maybe. I don't recall any major papers this year about efficiency in NeurIPS, ICLR, etc.
Nevertheless I think my point on "the wider Deep Learning community" not focusing on efficiency is correct.