The two properties you need for a siphon are the cohesion of the liquid (and this is true for the regular water siphon) and gravity, with the latter being the key player.
...and the former still being essential because if you lose cohesion you have two separate columns and nothing flowing.
Atmospheric pressure is not needed.
Atmospheric pressure is what provides the cohesion in the normal case of the water siphon. It means you can in theory have a siphon that climbs up to 10 meters. The experiment in the video indeed does not use atmospheric pressure. It relies on surface tension, which is much weaker. Even if water didn't boil in a vacuum the siphon would work only work for a height around a mm or so (high high can you "pull" water using its surface tension?). The liquid used in the experiment has a much higher surface tension than water, which is shy the siphon works at all. That being said, I doubt it would work for much higher than what was shown in the experiment -- if it did, the experimenters would have shown us a more impressive siphon.