TorchResist: Open-Source Differentiable Resist Simulator

2025-02-06

AI Summary

Existing lithographic simulators suffer from significant limitations in photoresist modeling accuracy, differentiability, and integration capability, hindering the efficiency of process optimizations such as optical proximity correction (OPC). This paper introduces TorchResist—the first open-source, end-to-end differentiable, white-box photoresist simulator. Its core innovation lies in a lightweight, physically grounded photoresist response function with ≤20 interpretable parameters, derived from analytical physical models. TorchResist tightly integrates PyTorch’s automatic differentiation, GPU-accelerated parallel computation, and rigorous optical propagation modeling, enabling joint mask–resist gradient-based optimization. Experimental results demonstrate that TorchResist surpasses state-of-the-art commercial and academic simulators in both accuracy and speed. It fills a critical gap in open-source, differentiable lithographic simulation and has been successfully integrated into industrial-grade OPC toolchains.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Computer Vision: Learning & Optimization for CV
• Machine Learning: Hardware-aware ML

Application Category

• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Recent decades have witnessed remarkable advancements in artificial intelligence (AI), including large language models (LLMs), image and video generative models, and embodied AI systems. These advancements have led to an explosive increase in the demand for computational power, challenging the limits of Moore's Law. Optical lithography, a critical technology in semiconductor manufacturing, faces significant challenges due to its high costs. To address this, various lithography simulators have been developed. However, many of these simulators are limited by their inadequate photoresist modeling capabilities. This paper presents TorchResist, an open-source, differentiable photoresist simulator.TorchResist employs an analytical approach to model the photoresist process, functioning as a white-box system with at most twenty interpretable parameters. Leveraging modern differentiable programming techniques and parallel computing on GPUs, TorchResist enables seamless co-optimization with other tools across multiple related tasks. Our experimental results demonstrate that TorchResist achieves superior accuracy and efficiency compared to existing solutions. The source code is publicly available.

Problem

• Develops open-source photoresist simulator
• Improves photoresist modeling accuracy
• Enables efficient semiconductor lithography optimization

Innovation

• Open-source photoresist simulator
• Differentiable programming techniques
• GPU-accelerated parallel computing