Super obnoxious when people act like they have the monopoly on evidence-based practice and don't even know how to fully consider the scope of the evidence base. The research on gestalt language processing and how to support language learning for autistic people is EVERYWHERE. Like, I'd be happy to show you where in the research it says that many children communicate with large chunks of language (echolalia). Where it says that effective therapy is both relevant and engaging to the client. Where it says that modeling language to fit a variety of communication functions and creating naturalistic communication opportunities is beneficial. Beyond peer-reviewed research, I'll throw in my own clinical experiences and knowledge of autistic perspectives, since clinical experiences and client perspectives are also two entire sides of the EBP triangle (which I've only ever seen depicted as an equilateral).
Now show me where it says that decontextualized language intervention is generalizable enough to support functional communication. Where it says that prompt-based "teaching" methodologies don't frequently create prompt-dependence that precludes the ability to use the skill functionally. Where it says that teaching children who communicate primarily through echolalia a series of single words and rote phrases is effective. I'll wait. In the meantime, consider that actual nuanced understanding of the body of evidence with regard to language development and autism is perhaps more significant than your keyword search.