Here’s some stuff I put together at one point or another. It might be useful for you, so email me for a copy.
EACL 2009 data
Zipped tarball (540K) containing the data used in our 2009 paper with human annotations. Contains 299 preprocessed and annotated XML documents plus a readme. If used, please cite Kastner and Monz (2009).
Database of unambiguous Hebrew verbs
A database of Modern Hebrew verbs that have no non-verb homographs, complete with root, binyan, gloss and other information. I needed this for a neurolinguistic experiment and it doesn’t look pretty enough to be released into the wild yet, but get in touch if you’d like the data.