Challenge: Academic documents are packed with texts, equations, tables, and figures, posing challenges for accurate OCR results.
Approach: They propose a model that integrates location guiding into the transformer architecture during autoregression.
Outcome: The proposed model outperforms existing methods on an original large-scale dataset comprising 53M text-location pairs from 89K academic document pages.

Similar Papers

PEaCE: A Chemistry-Oriented Dataset for Optical Character Recognition on Scientific Documents (2024.lrec-main)

Copied to clipboard

Challenge: Existing open-source OCR models focus on scientific texts or generic printed English . Nougat is unable to parse tables in PubMed articles .
Approach: They propose to train OCR models for scientific or generic printed English . Nougat is a popular tool for parsing academic documents, but unable to parse PubMed tables .
Outcome: The proposed models perform better when trained on real-world records than those trained on synthetic records.
Efficient OCR for Building a Diverse Digital History (2024.acl-long)

Copied to clipboard

Challenge: Current optical character recognition (OCR) systems are poorly extensible to low-resource document collections, as learning a language-vision model requires extensive labeled sequences and compute.
Approach: They propose to model optical character recognition as a character level image retrieval problem using a contrastively trained vision encoder.
Outcome: The proposed model is more sample efficient and extensible than existing architectures, enabling accurate OCR in settings where existing solutions fail.
Building OCR/NER Test Collections (2020.lrec-1)

Copied to clipboard

Challenge: Named entity recognition (NER) is the automatic recognition of spans of text as name mentions.
Approach: They propose a method for annotating named entities on transcribed text . the transcriptions are all that is needed to evaluate the performance of OCR .
Outcome: The proposed collection can be used to evaluate OCR and NER on transcribed text without re-annotation . the transcriptions are all that is needed to evaluate the performance of OCR .
Correction of OCR Word Segmentation Errors in Articles from the ACL Collection through Neural Machine Translation Methods (L18-1)

Copied to clipboard

Challenge: Optical Character Recognition (OCR) can produce a range of errors depending on the quality of the original document.
Approach: They applied a sequence-to-sequence machine translation system to correct word-single-word OCR errors in scientific texts from the ACL collection with an estimated precision and recall above 0.95 on test data.
Outcome: The proposed system corrects word-segmentation OCR errors with an estimated precision and recall above 0.95 on test data.
Evaluating Transformers for OCR Post-Correction in Early Modern Dutch Theatre (2025.coling-main)

Copied to clipboard

Challenge: a new study examines the effectiveness of two types of transformer models for OCR post-correction in early modern Dutch plays.
Approach: They propose to use large generative models and sequence-to-sequence models for OCR post-correction in early modern Dutch plays.
Outcome: The proposed model outperforms generative models on the OCR post-correction task . the model outpersforms the model with the lowest error rate on the historical English dataset .
Beyond Atomic Characters: Glyph-Aware Sub-character Alignment for Low-Resource Multilingual OCR (2026.acl-long)

Copied to clipboard

Challenge: Low-resource multilingual OCR models struggle with complex script structures and data scarcity.
Approach: They propose a framework for multilingual character recognition that integrates visual and linguistic backbones with a novel glyph-aware interface.
Outcome: The proposed framework improves on high-resolution visual and language backbones with glyph-aware interface.
Let’s Fuse Step by Step: A Generative Fusion Decoding Algorithm with LLMs for Robust and Instruction-Aware ASR and OCR (2025.findings-acl)

Copied to clipboard

Challenge: Various fusion strategies have been explored for integration of large language models into multi-modal systems.
Approach: They propose a framework for deep fusion decoding that integrates large language models into cross-modal text recognition systems.
Outcome: The proposed framework surpasses cascaded methods in English and Mandarin, and significantly reduces WERs by 17.7%.
EfficientOCR: An Extensible, Open-Source Package for Efficiently Digitizing World Knowledge (2023.emnlp-demo)

Copied to clipboard

Challenge: Existing OCR engines fail to provide accurate, cost-effective and sample-efficient character recognition for public domain documents.
Approach: EffOCR is an open-source optical character recognition package that is accurate, cheap to deploy and sample efficient to customize to novel collections, languages, and character sets.
Outcome: EffOCR model trains character retrieval problem and scales to novel collections, languages, and character sets.
Unsupervised Multi-View Post-OCR Error Correction With Language Models (2021.emnlp-main)

Copied to clipboard

Challenge: Prior work used text generation techniques or redundancy in similar passages for OCR error correction, which is not appropriate in cases of low corpus redundancies or weak document contextual information.
Approach: They propose to use a pretrained language model to reconcile different OCR views in unsupervised way so that their combination contains fewer errors than each individual view.
Outcome: The proposed model can reconcile multiple OCR views so that their combined version contains fewer errors than the best OCR view.
Quantifying Character Similarity with Vision Transformers (2023.emnlp-main)

Copied to clipboard

Challenge: Off-the-shelf string matching methods are widely used to link entities across datasets, but they are not available for many settings.
Approach: They propose to use augmented digital fonts to measure character substitution costs for OCR’ed documents by using vision transformers.
Outcome: The proposed method significantly improves record linkage compared to other string matching methods.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations