Is there any data augmentation during the training of slat flow matching models?

Thank you for your work!
When I'm testing the pretrained slat flow matching models, I found that this model is super robust to the generated coords (from the ss flow matching model). No matter the generated coords are with many false voxels/ small scaled geometry, it will accurately identify the correct subset and generate coherent slat feature for these voxels.
But I cannot find any scaling/ noise injection within the data process pipeline or dataset, which means the model have only seen the 3D assets in the normalized scale with clean coords, so I don't understand how does the slat flow matching model achieves such robustness.
The following pictures are original generated result from trellis/ coords visualization without slat feature(demonstrating the noise coords)/ rescaled coords before stage 2 generation.

No files, tests, or entry points are named. Trace the slat flow matching training data pipeline and dataset handling to determine whether scaling or noise augmentation is applied, then document the finding and the source of the model's observed robustness.