Is there a database somewhere of the review literature (especially the most cited papers) in theoretical physics? I know about Spires but unfortunately it is not working. When I open a page it says "Your search did not match any records. Please try again." Does anyone have this problem and what are the alternatives? I know the string wiki website which has a big collection of excellent review papers in string theory but it does not contain reviews in the standard model and other areas of theoretical physics.
The ADS database contains 11 million records of publications in astronomy, physics and geophysics, as well as arXiv preprints.
SPIRES has been replaced by INSPIRE as of a few weeks ago. There is a link to HEP reviews on the sidebar there, or you can probably try the same searches that SPIRES would do to find review papers in a particular category.
First of all, there are journals publishing a lot of review papers : Physics Reports, Reviews of Modern Physics, Living Reviews in Relativity, etc., so you may search on their web sites.
As for a database, the only one that comes to mind so far is the Net Advance of Physics.
Try Microsoft Academic Search , it has huge section devoted to Physics and you can search by string query or authors and pay attention to review journals, where articles were published, as @just-learning suggested.
what about the arxiv? You can find a lot of papers even before the are published and also search by research area (hep-ph, th-ph,...) arxiv
Google schoolar might help too.
I believe there is no complete database that would provide you with a reference to an authoritative review on every chosen topic.
There are obviously journals mentioned by user just-learning that often provide high-quality reviews from people at the frontier of their area. These journals in which the most authoritative reviews independent of the field are usually published are: Reviews of modern physics, Physical review letters and the Physics reports-review section of Physics letters. As for a focus more on experimental progress, excellent reviews are also found in Science and Nature.
If an article in these journal has a very general title or even the words "report" or "review" included, it is most likely the thing you are looking for.
As to the database - even if it existed, I would not use it. When I want to probe a topic I don't know, I type the keywords into an article search tool such as Web of Science or Google scholar and sort the results by citations (Web of Science tends to be more complete, but you need a university login). Amongst the first few of the results, the ones with the general titles will usually be the reviews "everyone knows", implicitly uses their assumptions, notation and copies the typos in their computations.
For example, type "Supersymmetry" into the search - Hans Nilles review is the thing you want. Or you want to know about the so-called "f(R) theories of gravity" - in both search tools you get immediately what you need.
For broader fields and for older reviews this usually works, but for reviews of recent developments in a more narrow field, you will have worse luck. Both the searches allow you to find articles only from a certain date but you will usually just find articles reviewing the most recent problems or lines of research - not an overview of the overall paradigm shift.
Furthermore, if even written, these "minor paradigm shift reviews" are very often included as extended introductions of articles on specific topics. It is hard to imagine how a database of such more or less minor reviews could even be organized. In this finer case, there are usually just a few tens of scientists who actually can identify a good recent review and they, as far as I know, don't create databases in any unified manner. Instead, they mention that this and this article had a nice review in their own articles. I am afraid this is where the detective work kicks in - you have to read at least the introductions of the most influential articles and trace the smaller reviews through references.