As a general rule/observation, the solubility of gases in the liquid will increase with increase in the temperature until it reaches the boiling point. This applies to liquid metal also. Look at this paper:

Article Solubility of hydrogen in liquid aluminium

The largest benefit of degassing is that entrained bifilms suspended within the melt float to the surface with the additional buoyancy provided by the argon gas. The equilibrium hydrogen solubility does not matter, in my opinion, because hydrogen can remain in solution in a supersaturated aluminum matrix by residing in vacancies. Therefore, my recommendation is to degas at the pouring temperature, right before pouring, as long as no additional entrainment damage occurs during degasing.

The optimal temperature range is highly dependent on the type of the degassing method. Besides hydrogen solubility, which is higher at elevated temperatures, it is important, that the oxidation rate of the melt also increases with temperature, which can cause accelerated loss of alloying elements which are prone to oxidation. Besides that, if you are using N2 as purging gas, the kinetics of the reaction between Al and N is also temperature-dependent. From this point of view, lower melt temperature is advantageous. On the other hand, if a cleaning flux blend is applied during the degassing treatment, higher melt temperature provides better fluidity for the molten flux inside the melt, which in turn, removes inclusions more efficiently. I agree with Prof. Murat Tiryakioḡlu regarding the importance of bifilm removal, however, we found that the number of small-sized bifilms can be increased during rotary degassing with N2 gas.

You ask about preference, but say nothing about the target preference criteria. In these conditions, nothing definite can be said. But I think it is possible not to ask the experts, but simply to find out by experiment.