Hey, are you interested in generating handwritten fonts? I came across a paper from Huawei that was published in CVPR 22 which uses GANs - <a href="https://openaccess.thecvf.com/content/CVPR2022/papers/Kong_Look_Closer_To_Supervise_Better_One-Shot_Font_Generation_via_Component-Based_CVPR_2022_paper.pdf" rel="nofollow">https://openaccess.thecvf.com/content/CVPR2022/papers/Kong_L...</a><p>If you are interested in LLMs, you can look into this paper from ByteDance (also from CVPR 22) - <a href="https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_XMP-Font_Self-Supervised_Cross-Modality_Pre-Training_for_Few-Shot_Font_Generation_CVPR_2022_paper.pdf" rel="nofollow">https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_XM...</a><p>The second paper seems to be more specific to Chinese fonts.
Not specifically LLM, but it looks like generating fonts with GAN is the subject ot this work: <a href="https://arxiv.org/abs/2203.10348" rel="nofollow">https://arxiv.org/abs/2203.10348</a><p>"Our goal is to generate fonts with specific impressions, by training a generative adversarial network with a font dataset with impression labels". The PDF contains interesting examples of Latin-based fonts generated by suggesting different impressions.