mlsanityData to 14 Oct 2026

Gen-Fab: A Variation-Aware Generative Model for Predicting Fabrication Variations in Nanophotonic Devices

arXiv 2603.11505 ↗First posted 12 Mar 2026cs.CV cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 21 of its 42 references that are in the map. First posted on arXiv 12 Mar 2026.

Builds on

21 of 42 references.

References · 21

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  4. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  5. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  6. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,527
  7. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  8. 2015Variational Inference with Normalizing Flowscited by 2,170
  9. 2014Conditional Generative Adversarial Netscited by 2,162
  10. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,323
  11. 2019Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shiftcited by 1,287
  12. 2016Instance Normalization: The Missing Ingredient for Fast Stylizationcited by 1,198
9 more
  1. 2015Empirical Evaluation of Rectified Activations in Convolutional Networkcited by 519
  2. 2017VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learningcited by 200
  3. 2019Seeing What a GAN Cannot Generatecited by 148
  4. 2018Patch-Based Image Inpainting with Generative Adversarial Networkscited by 70
  5. 2018Approximability of Discriminators Implies Diversity in GANscited by 28
  6. 2020BUSU-Net: An Ensemble U-Net Framework for Medical Image Segmentationcited by 2
  7. 2023Improving Fabrication Fidelity of Integrated Nanophotonic Devices Using Deep Learningcited by 2
  8. 2022Dynamic-Pix2Pix: Noise Injected cGAN for Modeling Input and Target Domain Joint Distributions with Limited Training Datacited by 1
  9. 2024SEMU-Net: A Segmentation-based Corrector for Fabrication Process Variations of Nanophotonics with Microscopic Imagescited by 1

Built on by

Not cited yet.