Abstract: Described are embodiments for generating a post-optical proximity correction (OPC) result for a mask using a target pattern and reference layer patterns. Images of the target pattern and reference layers are provided as an input to a machine learning (ML) model to generate a post-OPC image. The images may be input separately or combined into a composite image (e.g., using a linear function) and input to the ML model. The images are rendered from pattern data. For example, a target pattern image is rendered from a target pattern to be printed on a substrate, and a reference layer image such as dummy pattern image is rendered from dummy pattern. The ML model is trained to generate the post-OPC image using multiple images associated with target patterns and reference layers, and using a reference post-OPC image of the target pattern. The post-OPC image may be used to generate a post-OPC mask.

Abstract: A method for determining a mask pattern and a method for training a machine learning model. The method for determining a mask pattern includes obtaining, via executing a model using a target pattern to be printed on a substrate as an input pattern, a post optical proximity correction (post-OPC) pattern; determining, based on the post-OPC pattern, a simulated pattern that will be printed on the substrate; and determining the mask pattern based on a difference between the simulated pattern and the target pattern. The determining of the mask pattern includes modifying, based on the difference, the input pattern inputted to the model such that the difference is reduced; and executing, using the modified input pattern, the model to generate a modified post-OPC pattern from which the mask pattern can be derived.