Content modeling using latent permutations

We present a novel Bayesian topic model for learning discourse-level document struc-ture. Our model leverages insights from discourse theory to constrain latent topic assign-ments in a way that reflects the underlying organization of document topics. We propose a global model in which both topic selection and ordering are biased to be similar across a collection of related documents. We show that this space of orderings can be effectively rep-resented using a distribution over permutations called the Generalized Mallows Model. We apply our method to three complementary discourse-level tasks: cross-document alignment, document segmentation, and information ordering. Our experiments show that incorpo-rating our permutation-based model in these applications yields substantial improvements in performance over previously proposed methods.1 1