MaskOpt

MaskOpt is a large-scale benchmark dataset containing 104,714 metal-layer tiles and 121,952 via-layer tiles from real integrated circuit designs at the 45 nm node, used to evaluate deep learning approaches for cell- and context-aware mask optimization in optical lithography.

Papers using MaskOpt (1)

MaskOpt: A Large-Scale Mask Optimization Dataset to Advance AI in Integrated Circuit Manufacturing2025