For significance, we are encouraged that all reviewers

We thank reviewers for their thoughtful and positive feedback. We will answer R3's concern about the difference between our model and the conventional We will answer R2's main concern We address some specific comments below and will incorporate all the feedbacks in the revised paper . For R1 - We again appreciate your positive feedback. ": We would like to consider a variant BBP) can naturally lead to such steps. ": No, each model is allowed to have unlimited number For our MCMC inference, the number of active blocks is allowed to be variable order.