"Removes the requirement for an unmodified control sample.."

[anaesco]

Hi there, I have a question on using samples that are not unmodified, for comparison in xPore.

So in the paper "Identification of differential RNA modifications from nanopore direct RNA sequencing with xPore," in the intro it is stated that:
"xPore removes the requirement for an unmodified control sample and enables simultaneous profiling of differential transcript expression and modification across conditions without any additional experiments."

Does this mean that when running xPore, I can have one sample being treated with siRNA(50% KO), and the other with scrambled RNA(0% KO), and I would be able to compare modifications between the two?

Based on a post I saw on this github, I read that if I wanted to not use a control sample then it would be better to run m6Anet.
but m6Anet is an MIL program, so wouldn't I have to train it with an unmodified sample as well?

I hope this makes sense as I am quite confused as to which tool is most compatible with my sequencing data. And thank you for any guidance you can give me!