Abstract: A method for generating a pre-trained language model, includes: obtaining sample files; obtaining typography structure information and text information of the sample files by parsing the sample files; obtaining a plurality of task models of a pre-trained language model; obtaining a trained pre-trained language model by jointly training the pre-trained language model and the plurality of task models according to the typography structure information and the text information; and generating a target pre-trained language model by fine-tuning the trained pre-trained language model according to the typography structure information and the text information.

Abstract: Example implementations relate to halftone image creation. An example non-transitory machine-readable medium can include instructions executable to determine a highlight core shape and a shadow core shape of a microcell within a supercell. The instructions can be executable to determine growth sequences for a plurality of pixels within the highlight core and the shadow core and between the microcell and other microcells within the supercell, divide each of the plurality of pixels into a plurality of subpixels, and create a halftone image for an unequal resolution printing device using a constrained direct binary search model and based on the highlight core shape, shadow core shape, growth sequences, and the plurality of subpixels.

Abstract: Example implementations relate to halftone image creation. An example non-transitory machine-readable medium can include instructions executable to determine a highlight core shape and a shadow core shape of a microcell within a supercell. The instructions can be executable to determine growth sequences for a plurality of pixels within the highlight core and the shadow core and between the microcell and other microcells within the supercell, divide each of the plurality of pixels into a plurality of subpixels, and create a halftone image for an unequal resolution printing device using a constrained direct binary search model and based on the highlight core shape, shadow core shape, growth sequences, and the plurality of subpixels.