{"ID":2876956,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.21098","arxiv_id":"2508.21098","title":"TrInk: Ink Generation with Transformer Network","abstract":"In this paper, we propose TrInk, a Transformer-based model for ink generation, which effectively captures global dependencies. To better facilitate the alignment between the input text and generated stroke points, we introduce scaled positional embeddings and a Gaussian memory mask in the cross-attention module. Additionally, we design both subjective and objective evaluation pipelines to comprehensively assess the legibility and style consistency of the generated handwriting. Experiments demonstrate that our Transformer-based model achieves a 35.56\\% reduction in character error rate (CER) and an 29.66% reduction in word error rate (WER) on the IAM-OnDB dataset compared to previous methods. We provide an demo page with handwriting samples from TrInk and baseline models at: https://akahello-a11y.github.io/trink-demo/","short_abstract":"In this paper, we propose TrInk, a Transformer-based model for ink generation, which effectively captures global dependencies. To better facilitate the alignment between the input text and generated stroke points, we introduce scaled positional embeddings and a Gaussian memory mask in the cross-attention module. Additi...","url_abs":"https://arxiv.org/abs/2508.21098","url_pdf":"https://arxiv.org/pdf/2508.21098v1","authors":"[\"Zezhong Jin\",\"Shubhang Desai\",\"Xu Chen\",\"Biyi Fang\",\"Zhuoyi Huang\",\"Zhe Li\",\"Chong-Xin Gan\",\"Xiao Tu\",\"Man-Wai Mak\",\"Yan Lu\",\"Shujie Liu\"]","published":"2025-08-28T01:44:15Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.AI\"]","methods":"[\"Transformer\"]","has_code":false}