I'm not sure if it makes it easier, but there are some differences between the high oxygen titanium alloy and titanium with some yttrium in it that might make it easier to separate?

Presumably when you melt the titanium the yttrium doesn't react, whereas the oxygen dissolved in the titanium alloy at room temperature will form titanium dioxide when it's heated (if I'm reading correctly). So maybe you could "just" separate the molten metal by density afterwards? I'm not sure this would work though. For one, you'd need to avoid re-introducing oxygen contamination, but I guess you could do it under a vacuum (yes "just" spin the molten metal at high speed in a vacuum)?

This would seem to me to beg the question of why not just grind up the titanium in a vacuum to remove the oxygen and then melt it down, so I might be missing something here.