For now, start with the LISA publications database, which contains, as of this writing 26 publications related to deep architectures (this includes journal articles and conference proceedings but also theses, tech reports, Snowbird abstracts etc.). It is a good place to visit if you want to know about the latest & greatest work that comes out of our lab.
There is also the reading list compiled by the ICML 2009 Workshop on Learning Feature Hierarchies organizers, which provides a pretty good overview of the literature.
The rest is work in progress so stay tuned.
"A fast learning algorithm for deep belief nets" by Geoffrey E. Hinton, Simon Osindero, and Yee-Whye Teh. This is the paper that started it all. It presents a clever way of constructing a hierarchy of Restricted Boltzmann Machines