A few brief notes:<p>Not at all sure the recommendations system papers should be grouped with QA papers. The techniques used aren't that closely related.<p>FastText(.zip) continues to be a weird project.<p><i>LEARNING A NATURAL LANGUAGE INTERFACE WITH NEURAL PROGRAMMER</i> - yeah, that's something. The neural programmer and related stuff seems like something out of sci-fi, even more so than normal neural networks.
I provided a write-up of interesting ICLR submissions, primarily for NLP, but have also provided a small snippet explaining why I thought the paper was important or how it fit with other recent relevant papers.<p><a href="http://smerity.com/articles/2016/iclr_2017_submissions.html" rel="nofollow">http://smerity.com/articles/2016/iclr_2017_submissions.html</a><p>I'd also strongly recommend people read through all the submissions themselves as every person seems to select a different set of papers depending on what they're interested in. My write-up has very few papers discussing CNNs for example as I'm focused on RNNs for the most part.<p><a href="http://openreview.net/group?id=ICLR.cc/2017/conference" rel="nofollow">http://openreview.net/group?id=ICLR.cc/2017/conference</a>