Performance Gap Analysis between Latin and Arabic Scripts HTR

• URL: http://arxiv.org/abs/2606.18884v1
• Date: Wed, 17 Jun 2026 10:00:22 GMT
• Title: Performance Gap Analysis between Latin and Arabic Scripts HTR
• Abstract summary: We present a comprehensive study of Arabic and Latin scripts HTR using a unified CRNN model for line-level HTR.<n>Our results show the performance gap remains: it is large in low-resource settings, decreases with more data, but remains even at full scale, with a consistent difference of 5-7 CER points.
• Score: 5.227985502946548
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent studies have shown that handwritten text recognition (HTR) systems perform worse on Arabic-script datasets than on Latin-script data. However, the reasons for this gap are still not well understood due to the lack of controlled comparisons. In this work, we present a comprehensive study of Arabic and Latin scripts HTR using a unified CRNN model for line-level HTR across nine datasets (including KHATT (Arabic), Muharaf (Arabic), NUST-UHWR (Urdu), PHTD (Persian), IAM (English), READ-2016 (German), and others) and di ferent training sizes (K in {100, 500, 1000, 2000, ..., Kfull}). Our results show the performance gap remains: it is large in low-resource settings, decreases with more data, but remains even at full scale, with a consistent difference of 5-7 CER points. We show that annotation quality matters, as many datasets contain labeling errors. Cleaning reduces error rates and narrows the gap, but does not eliminate it. In addition, we find that a fixed number of training samples provides less effective coverage in Arabic due to higher visual variability, requiring more data to learn similar representations. We compare recognition across datasets in terms of the number of text lines and the number of characters, showing an equivalence trade-off. We compare character frequency distributions across scripts and show that Arabic is significantly more heavy-tailed than Latin. Our error analysis reveals that around 30 percent of substitution errors in Arabic datasets (e.g., KHATT) are caused by confusion between visually similar characters, compared to about 15 percent in Latin-script datasets such as IAM.

Related papers

• Cross-Language Learning within Arabic Script for Low-Resource HTR [5.227985502946548]
  Cross-script recognition is largely driven by script-level overlap rather than uniform accuracy improvements.<n>We conduct a controlled line-level study of cross-script joint training for Arabic-script HTR under low-resource regimes.<n>On Persian (PHTD), joint training achieves a Character Error Rate (CER) of 9.99, surpassing previously reported results.
  arXiv  Detail & Related papers  (2026-05-03T23:18:21Z)
• Group-Adaptive Threshold Optimization for Robust AI-Generated Text Detection [58.419940585826744]
  We introduce FairOPT, an algorithm for group-specific threshold optimization for probabilistic AI-text detectors.<n>We partitioned data into subgroups based on attributes (e.g., text length and writing style) and implemented FairOPT to learn decision thresholds for each group to reduce discrepancy.<n>Our framework paves the way for more robust classification in AI-generated content detection via post-processing.
  arXiv  Detail & Related papers  (2025-02-06T21:58:48Z)
• Unlikelihood Tuning on Negative Samples Amazingly Improves Zero-Shot Translation [79.96416609433724]
  Zero-shot translation (ZST) aims to translate between unseen language pairs in training data. The common practice to guide the zero-shot language mapping during inference is to deliberately insert the source and target language IDs. Recent studies have shown that language IDs sometimes fail to navigate the ZST task, making them suffer from the off-target problem.
  arXiv  Detail & Related papers  (2023-09-28T17:02:36Z)
• NusaWrites: Constructing High-Quality Corpora for Underrepresented and Extremely Low-Resource Languages [54.808217147579036]
  We conduct a case study on Indonesian local languages. We compare the effectiveness of online scraping, human translation, and paragraph writing by native speakers in constructing datasets. Our findings demonstrate that datasets generated through paragraph writing by native speakers exhibit superior quality in terms of lexical diversity and cultural content.
  arXiv  Detail & Related papers  (2023-09-19T14:42:33Z)
• MRN: Multiplexed Routing Network for Incremental Multilingual Text Recognition [56.408324994409405]
  Multiplexed routing network (MRN) trains a recognizer for each language that is currently seen. MRN effectively reduces the reliance on older data and better fights against catastrophic forgetting. It outperforms existing general-purpose IL methods by large margins.
  arXiv  Detail & Related papers  (2023-05-24T06:03:34Z)
• Context-Gloss Augmentation for Improving Arabic Target Sense Verification [1.2891210250935146]
  The most common semantically-labeled dataset for Arabic is the ArabGlossBERT. This paper presents an enrichment to the ArabGlossBERT dataset, by augmenting it using machine back-translation. We measure the impact of augmentation using different data configurations to fine-tune BERT on target sense verification (TSV) task.
  arXiv  Detail & Related papers  (2023-02-06T21:24:02Z)
• Towards Boosting the Accuracy of Non-Latin Scene Text Recognition [27.609596088151644]
  Scene-text recognition is remarkably better in Latin languages than the non-Latin languages. This paper examines the possible reasons for low accuracy by comparing English datasets with non-Latin languages.
  arXiv  Detail & Related papers  (2022-01-10T06:36:43Z)
• ChrEnTranslate: Cherokee-English Machine Translation Demo with Quality Estimation and Corrective Feedback [70.5469946314539]
  ChrEnTranslate is an online machine translation demonstration system for translation between English and an endangered language Cherokee. It supports both statistical and neural translation models as well as provides quality estimation to inform users of reliability.
  arXiv  Detail & Related papers  (2021-07-30T17:58:54Z)
• A Multitask Learning Approach for Diacritic Restoration [21.288912928687186]
  In many languages like Arabic, diacritics are used to specify pronunciations as well as meanings. Such diacritics are often omitted in written text, increasing the number of possible pronunciations and meanings for a word. We use Arabic as a case study since it has sufficient data resources for tasks that we consider in our joint modeling.
  arXiv  Detail & Related papers  (2020-06-07T01:20:40Z)
• Transliteration of Judeo-Arabic Texts into Arabic Script Using Recurrent Neural Networks [1.933681537640272]
  We train a model to automatically transliterate Judeo-Arabic texts into Arabic script. We employ a recurrent neural network (RNN), combined with the connectionist temporal classification (CTC) loss to deal with unequal input/output lengths. We obtain an improvement over the baseline 9.5% character error, achieving 2% error with our best configuration.
  arXiv  Detail & Related papers  (2020-04-23T18:03:41Z)
• A Sentence Cloze Dataset for Chinese Machine Reading Comprehension [64.07894249743767]
  We propose a new task called Sentence Cloze-style Machine Reading (SC-MRC) The proposed task aims to fill the right candidate sentence into the passage that has several blanks. We built a Chinese dataset called CMRC 2019 to evaluate the difficulty of the SC-MRC task.
  arXiv  Detail & Related papers  (2020-04-07T04:09:00Z)

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.