ScribeTokens: Fixed-Vocabulary Tokenization of Digital Ink

TL;DR

ScribeTokens introduces a fixed-vocabulary tokenization method for digital ink that improves recognition accuracy and training efficiency, outperforming previous vector-based and token-based methods in handwritten text recognition tasks.

Contribution

The paper proposes ScribeTokens, a fixed-vocabulary tokenization for digital ink that enhances recognition performance and training speed, and introduces a self-supervised pretraining strategy for better results.

Findings

ScribeTokens outperforms vector representations in recognition accuracy.

Pretraining with next-ink-token prediction improves convergence and accuracy.

Achieves state-of-the-art results on IAM and DeepWriting datasets.

Abstract

Digital ink -- the coordinate stream captured from stylus or touch input -- lacks a unified representation. Continuous vector representations produce long sequences and suffer from training instability, while existing token representations require large vocabularies, face out-of-vocabulary issues, and underperform vectors on recognition. We propose ScribeTokens, a tokenization that decomposes pen movement into unit pixel steps. Together with two pen-state tokens, this fixed 10-token base vocabulary suffices to represent any digital ink and enables aggressive BPE compression. On handwritten text generation, ScribeTokens dramatically outperforms vectors (17.33% vs. 70.29% CER), showing tokens are far more effective for generation. On recognition, ScribeTokens is the only token representation to outperform vectors without pretraining. We further introduce next-ink-token prediction as a self-supervised pretraining strategy, which consistently improves recognition across all token-based models and accelerates convergence by up to 83x. With pretraining, ScribeTokens achieves the best recognition results across all representations on both datasets (8.27% CER on IAM, 9.83% on DeepWriting).