Export Reviews, Discussions, Author Feedback and Meta-Reviews

Neural Information Processing Systems 

We thank the reviewers for their helpful comments and questions. We will implement all their valuable suggestions for improving the presentation and the readability of the text and the figures. For clarity, the practical contributions of this work can be summarized as follows. We present a general scheme for approximate inference in pairwise binary graphical models. The scheme requires constructing an "attractive" 2-cover of the base graph and performing belief propagation on this cover.