Well, Pandoc parses the following:

into a Paragraph Node, followed by a List Node, for which the LaTeX writer outputs the snippet above.

What I'm concerned with is the final output, but the whitespace added by the writer is significant in that it represents a new *LaTeX* Paragraph, and this is what adds the extra space in the generated output. This is confirmed by the AST JSON dump, and compilation of MWEs with the above snippets.

So the real question is: "Is there a way to get this input parsed in a way such that LaTeX will emit them as one paragraph unit?"