e836d813fd184325132fca8edcdfb40e-Reviews.html

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. Overview: This paper looks at the difficult problem of learning FST models of unaligned input and output sequences. This is an interesting problem and the approach appears to have merit; the main drawback of the paper is that some sections are very difficult to understand. In the abstract and introduction: the authors mention that the setting where sequences are not aligned is more realistic. I believe the authors, but examples of problems where unaligned sequences are the norm would be welcome.