9/2017: what's the topic?

Anybody read any neat papers lately?

Not exactly new (published 2014, revised 2017), but I’ve been intrigued by the posterior regularization ideas from Big Learning with Bayesian methods lately, though haven’t done anything serious with them yet.