I cringed at this point. Especially in mathematical fields where the abstract is in some sense an API -- it tells you what theorems you might use from this paper and how to apply them.