This is a very interesting idea. The author has used a networks idea
of dependence to analyze networks that might change over time. As
happens in time series, the data points (in this case networks) are
not independent over time because closer point might be more similar
than points that are farther apart in time. This seems like a natural
question to ask in network analysis, and it seems like it will be a
very useful paper for many researchers.

The literature review in section was very comprehensive, but it was a
little difficult to see how it all fit together. It seemed like a list
of lots of research that has been done with networks recently, but I'm
not sure how it's informing your research, or where the gap in the
research is that you will fill.

A way to write the literature review that might make it a little more
fun to read is that, instead of listing what each author did in his or
her paper, you could focus on the types of things that have been done
in the literature, and then say who did this. I got this technique
from the Global Communications Center when I asked them to edit my
first paper. They helped me structure the topics going from broadest
(networks) to narrowest (time series in networks). It might be useful
for you to go there and have them help you organize it in a way that's
pleasing to read.

The model looks good, but I get a little lost in the paragraphs
because they are a little long. A suggestion would be to make
paragraphs shorter and add a few sentences at the beginning of
paragraphs saying why I should care that you are doing this. This is a
way to create some tension so I want to keep reading. For example,
start a paragraph with, "Although the latent space network model (LSM)
introduced by Hoff et al. (2002) is characterized by the positions of
the nodes in a low-dimensional latent space, I will focus on doing XX
differently." Or something like that.

See additional comments in marked-up PDF