TrInk: Ink Generation with Transformer Network

• URL: http://arxiv.org/abs/2508.21098v1
• Date: Thu, 28 Aug 2025 01:44:15 GMT
• Title: TrInk: Ink Generation with Transformer Network
• Abstract summary: TrInk is a Transformer-based model for ink generation that captures global dependencies.<n>We introduce scaled positional embeddings and a Gaussian memory mask in the cross-attention module.<n> Experiments demonstrate that our model achieves a 35.56% reduction in character error rate (CER) and an 29.66% reduction in word error rate (WER)
• Score: 41.562705515254144
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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/

Related papers

• The Cursive Transformer [0.6138671548064355]
  We introduce a novel tokenization scheme that converts pen stroke offsets to polar coordinates, discretizes them into bins, and then turns them into sequences of tokens.<n>With just 3,500 handwritten words and a few simple data augmentations, we are able to train a model that can generate realistic cursive handwriting.
  arXiv  Detail & Related papers  (2025-03-31T03:22:27Z)
• Abstractive Summarization as Augmentation for Document-Level Event Detection [0.0]
  We bridge the performance gap between shallow and deep models on document-level event detection by using abstractive text summarization as an augmentation method. We use four decoding methods for text generation, namely beam search, top-k sampling, top-p sampling, and contrastive search. Our results show that using the document title offers 2.04% and 3.19% absolute improvement in macro F1-score for linear SVM and RoBERTa, respectively.
  arXiv  Detail & Related papers  (2023-05-29T11:28:26Z)
• Thutmose Tagger: Single-pass neural model for Inverse Text Normalization [76.87664008338317]
  Inverse text normalization (ITN) is an essential post-processing step in automatic speech recognition. We present a dataset preparation method based on the granular alignment of ITN examples. One-to-one correspondence between tags and input words improves the interpretability of the model's predictions.
  arXiv  Detail & Related papers  (2022-07-29T20:39:02Z)
• DPText-DETR: Towards Better Scene Text Detection with Dynamic Points in Transformer [94.35116535588332]
  Transformer-based methods, which predict polygon points or Bezier curve control points to localize texts, are quite popular in scene text detection. However, the used point label form implies the reading order of humans, which affects the robustness of Transformer model. We propose DPText-DETR, which directly uses point coordinates as queries and dynamically updates them between decoder layers.
  arXiv  Detail & Related papers  (2022-07-10T15:45:16Z)
• CoMER: Modeling Coverage for Transformer-based Handwritten Mathematical Expression Recognition [4.812445272764651]
  Transformer-based encoder-decoder architecture has recently made significant advances in recognizing handwritten mathematical expressions. Coverage information, which records the alignment information of the past steps, has proven effective in the RNN models. We propose CoMER, a model that adopts the coverage information in the transformer decoder.
  arXiv  Detail & Related papers  (2022-07-10T07:59:23Z)
• Hyperbolic Vision Transformers: Combining Improvements in Metric Learning [116.13290702262248]
  We propose a new hyperbolic-based model for metric learning. At the core of our method is a vision transformer with output embeddings mapped to hyperbolic space. We evaluate the proposed model with six different formulations on four datasets.
  arXiv  Detail & Related papers  (2022-03-21T09:48:23Z)
• HETFORMER: Heterogeneous Transformer with Sparse Attention for Long-Text Extractive Summarization [57.798070356553936]
  HETFORMER is a Transformer-based pre-trained model with multi-granularity sparse attentions for extractive summarization. Experiments on both single- and multi-document summarization tasks show that HETFORMER achieves state-of-the-art performance in Rouge F1.
  arXiv  Detail & Related papers  (2021-10-12T22:42:31Z)
• Diffusion models for Handwriting Generation [0.0]
  We propose a diffusion probabilistic model for handwriting generation. Our method does not require using any text-recognition based, writer-style based, or adversarial loss functions. Our model is able to incorporate writer stylistic features directly from image data, eliminating the need for user interaction during sampling.
  arXiv  Detail & Related papers  (2020-11-13T00:31:22Z)
• ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators [108.3381301768299]
  Masked language modeling (MLM) pre-training methods such as BERT corrupt the input by replacing some tokens with [MASK] and then train a model to reconstruct the original tokens. We propose a more sample-efficient pre-training task called replaced token detection.
  arXiv  Detail & Related papers  (2020-03-23T21:17:42Z)
• Paraphrase Generation with Latent Bag of Words [46.34571236204907]
  We propose a latent bag of words (BOW) model for paraphrase generation. We use source words to predict their neighbors and model the target BOW with a mixture of softmax. Our latent BOW model not only enhances the decoder, but also exhibits clear interpretability.
  arXiv  Detail & Related papers  (2020-01-07T09:22:58Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.