There's more to this than meets the eye
This is so plainly a ueless form of cypher (see numerous comments above) that it's totally implausable that a team of researchers in the field (with multiple papers behind them - these are not 1st year undergrads) would not know the weaknesses.
So what the hell is going on?
Like 'Disappointed' above, I thought this was a cunning way to hide messages in individual font shapes, which would be very cool, but is awkward in PDFs.
If it's not that, then what?
I wonder if they are really onto something much more clever, but had to yank the paper at the last minute at the request of some 'special' people. I've been there and you finish having to dash out something feeble to fill the gap.