Adoption of tier stacking (dual deck) leads to increasingly high aspect ratios and poses challenges in controlling overlay, tilt and misalignment for next generation 3D NAND devices manufacturing processes. In this work we address metrology challenges for asymmetry measurements. We show that Mueller measurement can separate overlay and tilt signals through distinct spectral response analyzed by machine learning with high measurement accuracy. We propose and demonstrate tilt and overlay accuracy improvement using nonlinear regression model combined with feeding forward critical dimension measurement results to the analysis of asymmetry signals to reduce errors caused by imperfect linearity of spectral response and structural variations. We demonstrate that spectroscopic Mueller matrix measurements, paired with advanced machine learning analysis, provide non-destructive and accurate asymmetry measurement for 3D NAND devices with high throughput and fast recipe creation.