Unitho: A Unified Multi-Task Framework for Computational Lithography
This addresses the need for reliable data foundations in computational lithography for the electronic design automation (EDA) industry, representing a domain-specific advancement.
The paper tackles the problem of isolated tasks in computational lithography by introducing Unitho, a unified multi-task large vision model based on Transformers, which achieves performance substantially surpassing academic baselines on mask generation, lithography simulation, and rule violation detection.
Reliable, generalizable data foundations are critical for enabling large-scale models in computational lithography. However, essential tasks-mask generation, rule violation detection, and layout optimization-are often handled in isolation, hindered by scarce datasets and limited modeling approaches. To address these challenges, we introduce Unitho, a unified multi-task large vision model built upon the Transformer architecture. Trained on a large-scale industrial lithography simulation dataset with hundreds of thousands of cases, Unitho supports end-to-end mask generation, lithography simulation, and rule violation detection. By enabling agile and high-fidelity lithography simulation, Unitho further facilitates the construction of robust data foundations for intelligent EDA. Experimental results validate its effectiveness and generalizability, with performance substantially surpassing academic baselines.