Saved in:
Bibliographic Details
Main Authors: Dev, Arundhathi, Zhan, Justin
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.28028
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917368576344064
author Dev, Arundhathi
Zhan, Justin
author_facet Dev, Arundhathi
Zhan, Justin
contents Optical character recognition remains critical infrastructure for document digitization, yet state-of-the-art performance is often restricted to well-resourced institutions by prohibitive computational barriers. End-to-end transformer architectures achieve strong accuracy but demand hundreds of GPU hours for domain adaptation, limiting accessibility for practitioners and digital humanities scholars. We present a modular detection-and-correction framework that achieves near-SOTA accuracy with single-GPU training. Our approach decouples lightweight visual character detection (domain-agnostic) from domain-specific linguistic correction using pretrained sequence models including T5, ByT5, and BART. By training the correctors entirely on synthetic noise, we enable annotation-free domain adaptation without requiring labeled target images. Evaluating across modern clean handwriting, cursive script, and historical documents, we identify a critical "Pareto frontier" in architecture selection: T5-Base excels on modern text with standard vocabulary, whereas ByT5-Base dominates on historical documents by reconstructing archaic spellings at the byte level. Our results demonstrate that this decoupled paradigm matches end-to-end transformer accuracy while reducing compute by approximately 95%, establishing a viable, resource-efficient alternative to monolithic OCR architectures.
format Preprint
id arxiv_https___arxiv_org_abs_2603_28028
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Efficient Domain Adaptation for Text Line Recognition via Decoupled Language Models
Dev, Arundhathi
Zhan, Justin
Computer Vision and Pattern Recognition
Machine Learning
Optical character recognition remains critical infrastructure for document digitization, yet state-of-the-art performance is often restricted to well-resourced institutions by prohibitive computational barriers. End-to-end transformer architectures achieve strong accuracy but demand hundreds of GPU hours for domain adaptation, limiting accessibility for practitioners and digital humanities scholars. We present a modular detection-and-correction framework that achieves near-SOTA accuracy with single-GPU training. Our approach decouples lightweight visual character detection (domain-agnostic) from domain-specific linguistic correction using pretrained sequence models including T5, ByT5, and BART. By training the correctors entirely on synthetic noise, we enable annotation-free domain adaptation without requiring labeled target images. Evaluating across modern clean handwriting, cursive script, and historical documents, we identify a critical "Pareto frontier" in architecture selection: T5-Base excels on modern text with standard vocabulary, whereas ByT5-Base dominates on historical documents by reconstructing archaic spellings at the byte level. Our results demonstrate that this decoupled paradigm matches end-to-end transformer accuracy while reducing compute by approximately 95%, establishing a viable, resource-efficient alternative to monolithic OCR architectures.
title Efficient Domain Adaptation for Text Line Recognition via Decoupled Language Models
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2603.28028