I'm not about to say that there's nothing new under the sun, but parsers are a really well-understood problem where 99.9% of people don't need frontier knowledge and wouldn't be in a position to use it anyway.
And I don't think that people doing research on parsers would ever rely on LLMs for precisely that reason. But we're not parser researchers right?
sam_lowry_•55m ago