Eh, I speak both English and Japanese and I would say that what counts as accurate translation is what is most proximate to that threshold of fundamental dissimilarity. Measuring translation accuracy for all languages the same way is more the problem here. There is no such thing as 1-1 to translation. It’s more like 1-(1+n) where n accounts for said distance. For languages with shared origins, n can be fairly small, while for those with entirely separate ones, it can be quite large.

That being said, Japanese to English translation in things like popular culture tends to take far too many liberties, I expect because the culture around Japanese translation in America has a very annoying, Orientalist bent, with people getting off on their “expertise” about a fake exoticism.