RefLAM: A Reference-Grounded Line Annotation Pipeline for Historical Arabic Manuscripts

• URL: http://arxiv.org/abs/2608.25140v1
• Date: Tue, 25 Aug 2026 20:46:11 GMT
• Title: RefLAM: A Reference-Grounded Line Annotation Pipeline for Historical Arabic Manuscripts
• Abstract summary: RefLAM is a pipeline converting manuscript page images and clean transcriptions into validated, line-level ground truth.<n>RefLAM couples a deep-learning page-segmentation model with a multimodal large language model (MLLM) for structured OCR.<n>We release AraMS-28k: 14 historical Arabic manuscript books, 3,043 pages, and 27,971 main-text and 629 margin-line annotations.
• Score: 1.583691353033079
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Existing approaches to building line-level Arabic handwritten-text-recognition (HTR) training data either rely on fully manual annotation, which does not scale, or on automatic OCR-to-reference alignment methods not yet extended to multi-script, two-zone (main-plus-margin) manuscript layouts with a provable correctness guarantee. We present RefLAM (Reference-grounded Line Annotation for Manuscripts), a pipeline converting manuscript page images and clean transcriptions into validated, line-level ground truth without sacrificing human oversight. RefLAM couples a deep-learning page-segmentation model with a multimodal large language model (MLLM) for structured OCR and a diacritic-agnostic fuzzy alignment engine that grounds each OCR line in a contiguous span of the reference text, with a character-level confidence score in $[0,100]$. A perfect score is provably equivalent to character-for-character identity of the normalised strings (the Confidence-100 rule), verified with no counterexample across the released corpus. A reviewer can thus trust a perfect score, confirming most lines at a glance rather than retyping them, so annotation becomes triaged, with attention concentrated on uncertain alignments. Across 7 fully page-validated books we measured a 75$\times$ throughput gain over manual annotation (3,000 vs. 40 lines/hr); applying the same guarantee to 7 further books, we retained 16,533 confidence-100 main-text lines within one week, excluding sub-100 lines rather than manually correcting them. Using RefLAM, we release AraMS-28k: 14 historical Arabic manuscript books, 3,043 pages, and 27,971 main-text and 629 margin-line annotations with bounding boxes, layout labels, and insertion anchors for 191 margin entries (30.4%). We also finetune Muharaf-pretrained baselines (including HATFormer) on AraMS-28k and report CER results confirming its practical utility for downstream HTR training.

Related papers

• A Corpus-Aligned Uthmani-to-Standard Quranic Word Mapping and a Deterministic Recitation Validator [0.0]
  Quranic text is distributed in two orthographic forms: the Uthmani script used in every printed mushaf, and the Standard (Imla'i) Arabic form that every mainstream Arabic NLP tool is built for.<n>We create a 2,290-pair, corpus-aligned Uthmani-to-Standard word mapping constructed by aligning the complete 6,236-verse Quran across both orthographic forms.<n>We build a deterministic, LLM-free Quranic recitation validator using a four-layer verse-matching search and word-error-rate-graded feedback.
  arXiv  Detail & Related papers  (2026-09-14T03:20:42Z)
• AraMS-28k: The Largest Publicly Released Line-Level Dataset of Historical Arabic Manuscripts with Margin and Insertion-Anchor Annotations [1.583691353033079]
  AraMS-28k is the largest publicly released line-level dataset of genuine historical Arabic manuscripts.<n>It comprises 14 books, 3,043 pages, and 28,600 annotated text lines.<n>The dataset was constructed with RefLAM, a reference-grounded annotation pipeline.
  arXiv  Detail & Related papers  (2026-08-27T10:16:41Z)
• Reading Order Inference for Complex Document Layouts [0.784573461502689]
  We present a method to evaluate the accuracy of non-modelled text-as-a-report.<n>We compare our method to the canonical Glossainaria layout in non-rectangular, non-modelled text.
  arXiv  Detail & Related papers  (2026-07-01T14:52:00Z)
• DocAtlas: Multilingual Document Understanding Across 80+ Languages [58.715440331861295]
  We introduce DocAtlas, a framework that constructs high-fidelity OCR datasets and benchmarks covering 82 languages.<n>Our dual pipelines, differential rendering of native DOCX documents and synthetic-based generation for right-to-left scripts produce precise structural annotations.
  arXiv  Detail & Related papers  (2026-05-12T18:09:38Z)
• How Far Is Document Parsing from Solved? PureDocBench: A Source-TraceableBenchmark across Clean, Degraded, and Real-World Settings [56.70440596502351]
  The past year has seen over 20 open document parsing models, yet benchmarks almost exclusively on OmniDocBench.<n>We present PureDocBench, a benchmark thatrenders document images from HTML/CSS, covering 10 domains, 66 pages, and 1,475 pages.
  arXiv  Detail & Related papers  (2026-05-08T09:30:31Z)
• Different Strokes for Different Folks: Writer Identification for Historical Arabic Manuscripts [0.5599792629509229]
  Handwritten Arabic manuscripts preserve the Arab world's intellectual and cultural heritage.<n>Using the Muharaf dataset of historical Arabic manuscripts, we evaluate writer identification from individual line images.<n>We propose a Convolutional Neural Network (CNN)-based model with attention mechanisms for closed-set writer identification.
  arXiv  Detail & Related papers  (2026-04-24T12:55:16Z)
• HoWToBench: Holistic Evaluation for LLM's Capability in Human-level Writing using Tree of Writing [84.84959282480641]
  We propose Tree-of-Writing (ToW) to resolve the implicit inconsistency often found when LLM-as-a-judge aggregates all sub-features in text evaluation.<n>We also present HowToBench, a large-scale Chinese writing benchmark encompassing 12 genres and 1302 instructions across three task categories.
  arXiv  Detail & Related papers  (2026-04-21T04:26:39Z)
• RATE: Reviewer Profiling and Annotation-free Training for Expertise Ranking in Peer Review Systems [6.083097040417168]
  We introduce LR-bench, a benchmark curated from 2024-2025 AI/NLP manuscripts with five-level self-assessed familiarity ratings.<n>We also propose RATE, a reviewer-centric ranking framework that distills each reviewer's recent publications into compact keyword-based profiles.<n>Our approach consistently achieves state-of-the-art performance, outperforming strong embedding baselines by a clear margin.
  arXiv  Detail & Related papers  (2026-01-27T14:13:46Z)
• CTC Transcription Alignment of the Bullinger Letters: Automatic Improvement of Annotation Quality [0.28617987842985093]
  Handwritten text recognition for historical documents remains challenging due to handwriting variability, degraded sources, and limited layout-aware annotations.<n>We introduce a self-training method based on a CTC alignment algorithm that matches full transcriptions to text line images using dynamic programming and model output probabilities trained with the CTC loss.<n>Our approach improves performance (e.g., by 1.1 percentage points CER with PyLaia) and increases alignment accuracy.
  arXiv  Detail & Related papers  (2025-08-11T12:18:41Z)
• Unblocking Fine-Grained Evaluation of Detailed Captions: An Explaining AutoRater and Critic-and-Revise Pipeline [58.832237984587664]
  We develop VNLI-Critique, a model for automated sentence-level factuality classification and critique generation.<n>We highlight three key applications: (1) VNLI-Critique demonstrates robust generalization, validated by state-of-the-art performance on the M-HalDetect benchmark; (2) The VNLI-Critique driven AutoRater for DOCCI-Critique provides reliable VLM rankings, showing excellent alignment with human factuality judgments; and (3) An innovative Critic-and-Revise pipeline, achieves substantial improvements in caption factuality.
  arXiv  Detail & Related papers  (2025-06-09T10:57:26Z)
• REInstruct: Building Instruction Data from Unlabeled Corpus [49.82314244648043]
  We propose REInstruct, a method to automatically build instruction data from an unlabeled corpus. By training Llama-7b on a combination of 3k seed data and 32k synthetic data from REInstruct, fine-tuned model achieves a 65.41% win rate on AlpacaEval leaderboard against text-davinci-003.
  arXiv  Detail & Related papers  (2024-08-20T09:05:03Z)
• End-to-End Page-Level Assessment of Handwritten Text Recognition [69.55992406968495]
  HTR systems increasingly face the end-to-end page-level transcription of a document. Standard metrics do not take into account the inconsistencies that might appear. We propose a two-fold evaluation, where the transcription accuracy and the RO goodness are considered separately.
  arXiv  Detail & Related papers  (2023-01-14T15:43:07Z)
• BN-HTRd: A Benchmark Dataset for Document Level Offline Bangla Handwritten Text Recognition (HTR) and Line Segmentation [0.0]
  We introduce a new dataset for offline Handwritten Text Recognition (HTR) from images of Bangla scripts comprising words, lines, and document-level annotations. The BN-HTRd dataset is based on the BBC Bangla News corpus, meant to act as ground truth texts. Our dataset includes 788 images of handwritten pages produced by approximately 150 different writers.
  arXiv  Detail & Related papers  (2022-05-29T22:56:26Z)

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.