The copy detection mechanism has been improved, now uses Google's DiffMatchPatch library instead of manually counting the amount of words appearance. Google's library is more efficient and also powerful, so now is possible to get the diff information (equals, added and removed).
List of changes: