Neural community centered language models relieve the sparsity trouble by the way they encode inputs. Word embedding layers make an ar