Paper

{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.11505v1</id>\n    <title>Gen-Fab: A Variation-Aware Generative Model for Predicting Fabrication Variations in Nanophotonic Devices</title>\n    <updated>2026-03-12T03:47:40Z</updated>\n    <link href='https://arxiv.org/abs/2603.11505v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.11505v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Silicon photonic devices often exhibit fabrication-induced variations such as over-etching, underetching, and corner rounding, which can significantly alter device performance. These variations are non-uniform and are influenced by feature size and shape. Accurate digital twins are therefore needed to predict the range of possible fabricated outcomes for a given design. In this paper, we introduce Gen-Fab, a conditional generative adversarial network (cGAN) based on Pix2Pix to predict and model uncertainty in photonic fabrication outcomes. The proposed method takes a design layout (in GDS format) as input and produces diverse high-resolution predictions similar to scanning electron microscope (SEM) images of fabricated devices, capturing the range of process variations at the nanometer scale. To enable one-to-many mapping, we inject a latent noise vector at the model bottleneck. We compare Gen-Fab against three baselines: (1) a deterministic U-Net predictor, (2) an inference-time Monte Carlo Dropout U-Net, and (3) an ensemble of varied U-Nets. Evaluations on an out-of-distribution dataset of fabricated photonic test structures demonstrate that Gen-Fab outperforms all baselines in both accuracy and uncertainty modeling. An additional distribution shift analysis further confirms its strong generalization to unseen fabrication geometries. Gen-Fab achieves the highest intersection-over-union (IoU) score of 89.8%, outperforming the deterministic U-Net (85.3%), the MC-Dropout U-Net (83.4%), and varying U-Nets (85.8%). It also better aligns with the distribution of real fabrication outcomes, achieving lower Kullback-Leibler divergence and Wasserstein distance.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <published>2026-03-12T03:47:40Z</published>\n    <arxiv:comment>Accepted and published in Structural and Multidisciplinary Optimization (2026)</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <arxiv:journal_ref>Structural and Multidisciplinary Optimization (2026)</arxiv:journal_ref>\n    <author>\n      <name>Rambod Azimi</name>\n    </author>\n    <author>\n      <name>Yuri Grinberg</name>\n    </author>\n    <author>\n      <name>Dan-Xia Xu</name>\n    </author>\n    <author>\n      <name>Odile Liboiron-Ladouceur</name>\n    </author>\n    <arxiv:doi>10.1007/s00158-026-04272-3</arxiv:doi>\n    <link href='https://doi.org/10.1007/s00158-026-04272-3' rel='related' title='doi'/>\n  </entry>"
}