So this is just a unusual case where the men just aren't as good as the women, rather than the women having an advantage.
I can't prove it, but I'm fairly sure this is the case.
I'm struggling to put this into words, without drawing diagrams and using phrases like 'small n samples', but I'll have a go. In events which are really popular (for instance marathon running) the difference between the fastest man and the fastest woman in a major race is probably fairly representative of the difference between fast men and fast women in general. However in these niche ultra-endurance events you can get cases where the best man is very good (for a man) whereas the best woman is excellent (for a woman). And the excellent woman beats the very good man.
The number of women doing ultra-endurance events is small (having the confidence to do something 'out there' does seem to be an issue). And the difference between world records will likely get smaller as more women do them. But there's no evidence that the difference is going to disappear, or change sign.
In the mid-pack (where I live...), women do often seem to out-perform men of similar raw ability, by executing (i.e. eating, drinking, pacing,) better. However at the sharp end, the best men don't all screw these things up.