A Scalable Framework for Automated NER Annotation Correction in Low-Resource Languages

Toqeer Ehsan; Thamar Solorio

doi:10.18653/v1/2026.findings-eacl.215

A Scalable Framework for Automated NER Annotation Correction in Low-Resource Languages

Toqeer Ehsan^*
, Thamar Solorio

^*Corresponding author for this work

Quantitative science and technology studies

Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

Abstract

Poor quality or noisy annotations in Named Entity Recognition (NER), as in any other NLP task, make it challenging to achieve state-of-the-art performance. In this paper, we present a multi-step framework to enhance the annotation quality of NER datasets by employing automated techniques. We propose a frequency-based iterative approach that leverages self-training and a dual-threshold mechanism to enhance inference confidence. Experimental evaluations on different NER datasets demonstrate significant improvements in NER performance with respect to the original datasets. This work further explores the potential of generative Large Language Models (LLMs) to perform NER for low-resource languages.

Original language	English
Title of host publication	Findings of the Association for Computational Linguistics: EACL 2026
Editors	V. Demberg, K. Inui, L. Marquez
Publisher	Association for Computational Linguistics (ACL)
Pages	4138-4151
Number of pages	14
ISBN (Electronic)	979-8-89176-386-9
DOIs	https://doi.org/10.18653/v1/2026.findings-eacl.215
Publication status	Published - 2026
MoE publication type	A4 Article in a conference publication
Event	19th Conference of the European Chapter of the Association for Computational Linguistics, Findings of EACL 2026 - Rabat, Morocco Duration: 24 Mar 2026 → 29 Mar 2026

Conference

Conference	19th Conference of the European Chapter of the Association for Computational Linguistics, Findings of EACL 2026
Country/Territory	Morocco
City	Rabat
Period	24/03/26 → 29/03/26

Access to Document

10.18653/v1/2026.findings-eacl.215Licence: CC BY

Cite this

@inproceedings{4da610b0a9204d53aa8a94d5b40c65eb,

title = "A Scalable Framework for Automated NER Annotation Correction in Low-Resource Languages",

abstract = "Poor quality or noisy annotations in Named Entity Recognition (NER), as in any other NLP task, make it challenging to achieve state-of-the-art performance. In this paper, we present a multi-step framework to enhance the annotation quality of NER datasets by employing automated techniques. We propose a frequency-based iterative approach that leverages self-training and a dual-threshold mechanism to enhance inference confidence. Experimental evaluations on different NER datasets demonstrate significant improvements in NER performance with respect to the original datasets. This work further explores the potential of generative Large Language Models (LLMs) to perform NER for low-resource languages.",

author = "Toqeer Ehsan and Thamar Solorio",

year = "2026",

doi = "10.18653/v1/2026.findings-eacl.215",

language = "English",

pages = "4138--4151",

editor = "V. Demberg and K. Inui and L. Marquez",

booktitle = "Findings of the Association for Computational Linguistics: EACL 2026",

publisher = "Association for Computational Linguistics (ACL)",

address = "United States",

note = "19th Conference of the European Chapter of the Association for Computational Linguistics, Findings of EACL 2026 ; Conference date: 24-03-2026 Through 29-03-2026",

}

Ehsan, T & Solorio, T 2026, A Scalable Framework for Automated NER Annotation Correction in Low-Resource Languages. in V Demberg, K Inui & L Marquez (eds), Findings of the Association for Computational Linguistics: EACL 2026. Association for Computational Linguistics (ACL), pp. 4138-4151, 19th Conference of the European Chapter of the Association for Computational Linguistics, Findings of EACL 2026, Rabat, Morocco, 24/03/26. https://doi.org/10.18653/v1/2026.findings-eacl.215

A Scalable Framework for Automated NER Annotation Correction in Low-Resource Languages. / Ehsan, Toqeer; Solorio, Thamar.
Findings of the Association for Computational Linguistics: EACL 2026. ed. / V. Demberg; K. Inui; L. Marquez. Association for Computational Linguistics (ACL), 2026. p. 4138-4151.

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - A Scalable Framework for Automated NER Annotation Correction in Low-Resource Languages

AU - Ehsan, Toqeer

AU - Solorio, Thamar

PY - 2026

Y1 - 2026

N2 - Poor quality or noisy annotations in Named Entity Recognition (NER), as in any other NLP task, make it challenging to achieve state-of-the-art performance. In this paper, we present a multi-step framework to enhance the annotation quality of NER datasets by employing automated techniques. We propose a frequency-based iterative approach that leverages self-training and a dual-threshold mechanism to enhance inference confidence. Experimental evaluations on different NER datasets demonstrate significant improvements in NER performance with respect to the original datasets. This work further explores the potential of generative Large Language Models (LLMs) to perform NER for low-resource languages.

AB - Poor quality or noisy annotations in Named Entity Recognition (NER), as in any other NLP task, make it challenging to achieve state-of-the-art performance. In this paper, we present a multi-step framework to enhance the annotation quality of NER datasets by employing automated techniques. We propose a frequency-based iterative approach that leverages self-training and a dual-threshold mechanism to enhance inference confidence. Experimental evaluations on different NER datasets demonstrate significant improvements in NER performance with respect to the original datasets. This work further explores the potential of generative Large Language Models (LLMs) to perform NER for low-resource languages.

UR - https://www.scopus.com/pages/publications/105039236578

U2 - 10.18653/v1/2026.findings-eacl.215

DO - 10.18653/v1/2026.findings-eacl.215

M3 - Conference article in proceedings

AN - SCOPUS:105039236578

SP - 4138

EP - 4151

BT - Findings of the Association for Computational Linguistics: EACL 2026

A2 - Demberg, V.

A2 - Inui, K.

A2 - Marquez, L.

PB - Association for Computational Linguistics (ACL)

T2 - 19th Conference of the European Chapter of the Association for Computational Linguistics, Findings of EACL 2026

Y2 - 24 March 2026 through 29 March 2026

ER -

A Scalable Framework for Automated NER Annotation Correction in Low-Resource Languages

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this