I'm probably sounding like a broken record but has the newly assembled regulator been cycled enough for it to be broken in? And has the hammer spring been adjusted to the knee of the velocity curve (95 - 97% of maximum velocity)?
Creep is caused by air weeping through microscopic imperfections in the regulator’s valve seat. Technically all regulators creep, it's just the extent to which it creeps that matters. If you're lucky, the surfaces mate really well right away and it won't creep a meaningful amount but many new regulators will benefit from being cycled a few hundred times to help the surfaces wear in against each other.
If it continues to creep, the only reliable solution I've found is to dress the two sides of the valve seat. One side is usually hard plastic (e.g. Delrin) and the other is usually metal (typ. brass or stainless steel). They need to be dead flat, parallel to each other, and polished free of any scratches. If all it needs is polishing, I’ve managed with nothing more than a thick piece of glass as a backer for wet/dry sandpaper, and a few careful strokes against 400, 600, 1200, and 2000 grit.
Regarding the hammer spring adjustment, if that isn't dialed in, the velocity will be particularly sensitive to pressure changes. When adjusted to the velocity knee as described above, small changes in pressure will not affect the velocity...just as a well-tuned unregulated PCP will hold a tight extreme spread over a wide pressure range.