LithoBench: Benchmarking AI Computational Lithography for Semiconductor Manufacturing

In addition to the data and data loaders, LithoBench also provides functionalities that can facilitate the development of and OpenILT [2], we implement the reference lithography simulation model as a PyTorch module, which can be used like a DNN layer. The GPU-based fast Fourier transform (FFT) can boost the speed of lithography simulation. PyTorch optimizers can be directly employed to optimize the masks according to ILT loss functions, significantly simplifying the development of ILT algorithms. To evaluate ILT results, LithoBench provides a simple interface to measure the L2 loss, PVB, EPE, and shots of the output masks. It also incorporates Python programs that can train and test the models mentioned in this paper. We provide the base classes of lithography simulation and mask optimization models. By inheriting the classes, users can easily build their own models that can be trained and tested by LithoBench, without the need of writing the code for data loading and evaluation. Fig. 1 shows a typical flow for training and evaluating an ILT model. The users only need to implement the model and the five functions, i.e. pretrain, train, save, load, run. We include a pretraining interface to support the commonly adopted two-stage training scheme. However, pretraining is optional since methods like DOINN do not use two-stage training. Similar interfaces are required for lithography simulation.