I have RNA seq data for the miRNA and mRNA. I wonder whether I should use same normalization methods for both of them or I can use different normalization methods for them. sometime in some papers I saw that, they used RPM for miRNA and RPKM for mRNA, but I don't know why.
Those two are not really different normalizations since the miRNA have a fixed length (or vary just a tiny bit) so normalizing by their length wouldn't do much.
In general you should be able to use different normalizations after all the normalization's role is to make data comparable among the same type.