About this document

A Comparative Study of Attention-Based Transformer and Deep CNN Architectures For Lithography Hotspot Detection in Advanced Process Nodes-1 by Prashanth Kumar A D is a document available to read on EtoBox.

This document presents a comparative study of two deep learning architectures, a Deep Convolutional Neural Network (CNN) and a Vision Transformer (ViT), for lithography hotspot detection in advanced semiconductor processes. The study highlights that while the CNN achieves higher accuracy (97.7%), the ViT demonstrates efficiency with fewer parameters, making it suitable for capturing long-range dependencies in complex layouts. The findings suggest that Transformer-based models may be advantageous for future

Author

Prashanth Kumar A D

Language

EN