I was able to work around the issue by creating a custom math function. I’m not aware of a way to take a vector as an input and then manipulate each element individually within a math expression node, nor does the math expression node support absolute operations on a vector either, but you can split the struct into the three floats, take them as separate inputs into the math function, and then recombine them into a vector struct using the vec() operation. The expression is:
vec((abs(x)), (abs(y)), (abs(z)))
