If you want to customise the output, you can build your own diff tool around Algorithm::Diff.

If order isn't important and each item appears no more than once per file,

my %seen;

++$seen{$_} while <$fh_in1>;
--$seen{$_} while <$fh_in2>;

for (keys(%seen)) {
   print($fh_out1 $_) if $seen{$_} > 0;
   print($fh_out2 $_) if $seen{$_} < 0;
}
[download]

If order isn't important and some items might appear more than once in a file,

my %seen1; ++$seen1{$_} while <$fh_in1>;
my %seen2; ++$seen2{$_} while <$fh_in2>;

print $fh_out1 grep !$seen2{$_}, keys %seen1;
print $fh_out2 grep !$seen1{$_}, keys %seen2;
[download]

Update: Added tons.