State recovery techniques for (M,L) algorithm