Machine learning based defect examination for semiconductor specimens

A computerized method of training a machine learning (ML) model usable for defect examination on a semiconductor specimen, the method comprising:

• obtaining a training set comprising one or more pairs of training images, each pair including an input image and a target image corresponding to the input image;
• and
• training the ML model alternately between two training modes using the training set, comprising, for a given pair of training images:
• in a stochastic mode, generating, by the ML model, a predicted reference image with a stochastic pattern variation (PV) from a PV distribution based on the input image, and optimizing the ML model based on the predicted reference image with respect to the target image, wherein the ML model is trained to treat all PVs within the PV distribution as normal and differentiate defective features therefrom, and wherein the PV distribution is learnt by the ML model based on PVs observed across the training set;
• and
• in a deterministic mode, generating, by the ML model, a predicted reference image with a predetermined PV selected from the PV distribution based on the target image, and optimizing the ML model based on the predicted reference image with respect to the target image itself, wherein the ML model is trained to generate a reference image sharing a similar PV as of an input image of the ML model.

A non-transitory computer readable storage medium tangibly embodying a program of instructions that, when executed by a computer, cause the computer to perform a method of defect examination on a semiconductor specimen, the method comprising:

• obtaining a runtime image of the semiconductor specimen;
• generating a reference image based on the runtime image using a machine learning (ML) model, wherein the ML model is previously trained alternately between two training modes using a training set, the ML model learning a pattern variation (PV) distribution based on PVs observed across the training set, the two training modes comprising:
• i) a stochastic mode where the ML model is configured to generate a predicted reference image with a stochastic pattern variation (PV) from a PV distribution, wherein the ML model is trained to treat all PVs within the PV distribution as normal and differentiate defective features therefrom, and
• ii) a deterministic mode where the ML model is configured to generate a predicted reference image with a predetermined PV selected from the PV distribution, wherein the ML model is trained to generate a reference image sharing a similar PV as of an input image of the ML model;
• and
• performing defect examination on the runtime image using the generated reference image.