This seems like a pretty vague question. Are you meaning between "all 3"? Or one of those topics. They are pretty broad topics. HCI is very applied, whereas the other two are highly linked and concrete/fundamental to CS.
If you need something between all three, I'd recommend looking into Image Processing.
I can't answer for the two first topics, but on Theory of Computation, since things do not move very fast in that domain, at least, in the openly available publications, the open problems of the 70's are still the open problems of today.
However, these are notable publications:
1. S. Cook's book (written with Phuong Nguyen): The Logical Foundations of Proof Complexity, is a major publication of 2014. It discusses the AC0 class, among other things.
2. The articles edited by N. Greenberg under the book title: Effective Mathematics of the Uncountable, discusses hyper-Turing formalisms and languages. The book is of 2014. Other such books are: Hypercomputation: Computing Beyond the Church-Turing Barrier by Apostolos Syropoulos (2008) and New Computational Paradigms: Changing Conceptions of What is Computable by S.B. Cooper, Benedikt Löwe and Andrea Sorbi (2007). A 2000 book is Computable Analysis by K. Weihrauch which discusses Type-2 Turing machines and Blum-Shub-Smale machines. The work follows the insights of mathematician S. Smale (most of his papers span the 70's to the 90's, so that may not be recent enough for you).
3. Complexity Theory of Real Functions by Ker-I Ko (2013) is a major book in complexity theory as well.
4. A paper that did not go unnoticed is Kane's proof that Unary Subset Sum is in Logspace using the generating function for partitions. This being said, generating functions for partition functions are said to date back from the days of Euler but were repopularized by Wiles' proof of the Fermat Last Theorem.
There are many other papers or books, but these are the ones that come to mind at the moment.