I can't remember a single academic poster, from my 'scientific years', that was (at all) well designed. All I can think of are walls and walls full of text (usually in the same font), and me not reading even 10% of them. But to be honest, most teams wouldn't have the budget for (or the tradition of) hiring designers to do them.
Posters usually need to cover a whole research project, and follow a certain constrained structure. They are mostly based on published work (or work that will be published soon), so they have to organise the information in a more or less standardised way (abstract/intro, problem, conclusion). I'm not sure about this, but I think most of them turn into publications sooner or later, so they are like a step in between.
I am curious about your question, though, I think it's a really good one. Doing a quick search I realised the best 'posters' I've seen are infographics, so you can probably incorporate some of those principles to the boring old poster. I can't really find any good examples from hard sciences, but this is how I would imagine a good poster could look like:
Edit: Found two new ones (the first is the winner for me)!