Gene name duplications

[Leonhard2000]

How does CPDB handle counts files where the same gene name appears more than once?
e.g.
ENSG00000000003
ENSG00000268895
ENSG00000175899
ENSG00000000003

Is the value of the first entry used or mean, sum, highest value?

This often appears with non_human data because several mice genes can have the same human orthologous gene.
Or should I make sure that my data has no gene name duplications?