Request for Pre-training and Fine-tuning Pipelines

[Areache]

Hi VIDARC Team,

I’m trying to reproduce both the pre-training and fine-tuning stages from your paper “VIDARC” (arXiv:2512.17661). While I understand the pre-training datasets are publicly available, I’m particularly interested in the training pipelines, scripts, hyperparameters, and dataset construction details for both stages.

Any guidance or pointers would be greatly appreciated. I will of course cite your paper if I use the resources.

Thanks!
Yujie