In particular, I am interested in the suggested common features of creole languages more or less grammaticalized by children.
Ray Jackendoff's "discussion note" in the latest issue of Language takes trenchant issue with UG, thus implying that its status is weak, but the sheer volume of recent citations Jackendoff cites implies that it is still the theory to beat.
The full citation: Jackendoff, Ray. 2011. Discussion note: What is the human language faculty? Two views. Language 87.3: 586-624. (self-archived pdf version)
There are some books and textbooks that are fairly recent which can give you a good place to start. One of the more recent ones is Understanding Minimalism, which I think is a good, pretty well-written introduction to some of the most recent stuff.
If you really want the bleeding edge, you will want to look at some of Chomsky's recent articles. You can find a list at his page at MIT.
Jackendoff is welcome to the group of those who doubt that anything like UG exists. All so-called linguistic universals are logical, biological, physiological, or neurological in nature. That is, there is no universal 'grammar'. Try Pathways of the Brain by Sydney Lamb.
See What counts as evidence in linguistics: The case of innateness, eds. Penke and Rosenbach. 2007. Amsterdam: John Benjamins. This collection of articles was originally published in Studies in Language.
Perhaps it's not that recent, but for reference - an exchange between Hauser, Chomsky and Finch on one side, and Jackendoff and Pinker on the other: Language Log.
In short: Chomsky's hypothesis is that UG contains nothing but recursion. Jackendoff and Pinker point out that our visual cognition is also recursive, so recursion can't belong to UG (unless we believe that these two types of recursion evolved completely independently - but why should we assume that?). If we take the Chomsky's premise about UG, and combine it with the J&G's observation, we end up with the conclusion that there is no UG.
Watch out about Evans & Levinson's paper. It's inaccurate about what Chomsky and others have said, and inaccurate about the languages they discuss. When you hunt down the references, they don't make anything like the claims that Evans & Levinson attribute to them.
These slides from a recent conference are pretty damning (and probably pretty accurate): "How (not) to uncover cross-linguistic variation" by Lisa Matthewson, who is a leading semanticist who specializes in fieldwork on Salish.
Another relevant article is "Mythomania? Methods and Morals from ‘The Myth of Language Universals’" by Daniel Harbour.