[amitport]

In this context, the rotation is for spreading energy and ensuring predictable coordinate distributions rather than diagonalization; it makes coordinate-wise quantization much more computationally efficient, though it throws away learnable structure.

[eecc]

ah ok, so intuitively it's like minimizing the error when replacing the values with a well-known distribution. So all you need to carry along is the rotation and the assumption that there is some amount of loss.