Do you actually know this to be the case, or are you guessing?

Have people done studies where (1) nobody is exemplified and all of the focus is on the math; (2) the mathematicians behind the most critical concepts that are being used are exemplified to some degree; or (3) you talk about mathematicians who "exemplify the social identities of the learners in your classroom"?

And with a large enough sample size, with sufficiently motivated teachers, etc., for the results to be meaningful?

There are reasons to be skeptical. Stereotype threat has been shown, repeatedly, to diminish attention and performance of students whose social identities are not culturally associated with expectations of mathematical adeptness. The default assumption should be that anything that is about identity is a net negative, because (a) it is completely irrelevant to the task, and (b) it risks evoking a known psychological effect that makes things worse.