Cross-Language Learning within Arabic Script for Low-Resource HTR

2026-05-03

AI Summary

This work addresses the performance degradation in handwritten text recognition for low-resource Arabic-script languages due to scarce annotated data by proposing a multilingual joint training approach that leverages their shared writing system. Through character-level transfer learning across Arabic, Urdu, and Persian, the study demonstrates that cross-lingual performance gains primarily stem from the transferability of shared characters rather than uniform improvements across all characters. Employing a CRNN architecture combined with character-level statistical analysis, the method achieves new state-of-the-art results under low-resource conditions: a character error rate (CER) of 9.99% on the Persian PHTD dataset and a significant reduction from 17.20% to 14.45% CER on the Urdu UNHD dataset, both outperforming monolingual baselines.

Abstract

Handwritten Text Recognition (HTR) under limited labeled data remains a challenging problem, particularly for Arabic-script languages. Although modern sequence-based recognizers perform well in high-resource settings, their accuracy degrades sharply as training data becomes scarce. Arabic-script languages share a common writing system with substantial character overlap, motivating cross-script training as a strategy to mitigate data scarcity. We performed experiments on Arabic, Urdu, and Persian scripts and achieved improvements over single-script baselines (new SotA especially for low-resource settings). A key finding of our experiments is that cross-script transfer is largely driven by script-level overlap rather than uniform accuracy improvements. Through a statistical character-level analysis we show that gains are structurally concentrated on characters shared across scripts, while language-specific characters exhibit limited or negative transfer. These findings provide insight into transfer dynamics in low-resource script families. Detailed results include: We conduct a controlled line-level study of cross-script joint training for Arabic-script HTR under low-resource regimes (number of samples K \in 100, 500, 1000 labeled lines) on Arabic (KHATT), Urdu (NUST-UHWR), and Persian (PHTD). A CRNN model is trained on the union of multiple related Arabic-script datasets and evaluated on individual target languages. On Persian (PHTD), joint training achieves a Character Error Rate (CER) of 9.99, surpassing previously reported results despite not using the full available training data. On an Urdu dataset (UNHD), joint training reduces CER from 17.20 to 14.45. Code and data splits are released to ensure reproducibility.1

Problem

• Handwritten Text Recognition
• Low-Resource
• Arabic Script
• Cross-Language Learning
• Data Scarcity

Innovation

• cross-script transfer
• low-resource HTR
• Arabic script
• joint training
• character-level analysis