TY - GEN

T1 - PEaCE

T2 - Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

AU - Zhang, Nan

AU - Heaton, Connor

AU - Okonsky, Sean Timothy

AU - Mitra, Prasenjit

AU - Toraman, Hilal Ezgi

N1 - Publisher Copyright: © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

PY - 2024

Y1 - 2024

N2 - Optical Character Recognition (OCR) is an established task with the objective of identifying the text present in an image. While many off-the-shelf OCR models exist, they are often trained for either scientific (e.g., formulae) or generic printed English text. Extracting text from chemistry publications requires an OCR model that is capable in both realms. Nougat, a recent tool, exhibits strong ability to parse academic documents, but is unable to parse tables in PubMed articles, which comprises a significant part of the academic community and is the focus of this work. To mitigate this gap, we present the Printed English and Chemical Equations (PEaCE) dataset, containing both synthetic and real-world records, and evaluate the efficacy of transformer-based OCR models when trained on this resource. Given that real-world records contain artifacts not present in synthetic records, we propose transformations that mimic such qualities. We perform a suite of experiments to explore the impact of patch size, multi-domain training, and our proposed transformations, ultimately finding that models with a small patch size trained on multiple domains using the proposed transformations yield the best performance. Our dataset and code is available at https://github.com/ZN1010/PEaCE.

AB - Optical Character Recognition (OCR) is an established task with the objective of identifying the text present in an image. While many off-the-shelf OCR models exist, they are often trained for either scientific (e.g., formulae) or generic printed English text. Extracting text from chemistry publications requires an OCR model that is capable in both realms. Nougat, a recent tool, exhibits strong ability to parse academic documents, but is unable to parse tables in PubMed articles, which comprises a significant part of the academic community and is the focus of this work. To mitigate this gap, we present the Printed English and Chemical Equations (PEaCE) dataset, containing both synthetic and real-world records, and evaluate the efficacy of transformer-based OCR models when trained on this resource. Given that real-world records contain artifacts not present in synthetic records, we propose transformations that mimic such qualities. We perform a suite of experiments to explore the impact of patch size, multi-domain training, and our proposed transformations, ultimately finding that models with a small patch size trained on multiple domains using the proposed transformations yield the best performance. Our dataset and code is available at https://github.com/ZN1010/PEaCE.

KW - Chemistry-Oriented Document Analysis

KW - Image to Text

KW - OCR Dataset

KW - Optical Character Recognition (OCR)

UR - http://www.scopus.com/inward/record.url?scp=85195974335&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85195974335

SP - 12679

EP - 12689

BT - Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation

A2 - Calzolari, Nicoletta

A2 - Kan, Min-Yen

A2 - Hoste, Veronique

A2 - Lenci, Alessandro

A2 - Sakti, Sakriani

A2 - Xue, Nianwen

Y2 - 20 May 2024 through 25 May 2024

ER -