Call for Papers

Call for Papers – AbjadNLP 2026

The 2nd Workshop on NLP for Languages Using Arabic Script
Co-located with EACL 2026, Rabat, Morocco

Following the successful inaugural AbjadNLP 2025 at COLING in Abu Dhabi, we are delighted to announce the second edition — AbjadNLP 2026, to be co-located with EACL 2026 in Rabat, Morocco. AbjadNLP is now officially part of SigArab, the ACL Special Interest Group on Arabic NLP, underscoring our commitment to sustained, community-driven research for languages using the Arabic script. We work closely with the Masakhane NLP community in Africa, with two of our organisers actively engaged in this network, ensuring strong African representation and collaboration.

The workshop is dedicated to advancing innovation and fostering deeper understanding in Natural Language Processing (NLP) for languages that use the Arabic script in any of its forms. Our primary focus is on:

  • Abjad languages, traditionally associated with Semitic scripts where consonants are explicitly represented.

  • Ajami languages, representing the adaptation of the Arabic script to write African languages in alphabetic form.

While the workshop name uses “Abjad” for simplicity, our scope extends to all languages written in Arabic, Perso-Arabic, and Ajami scripts — from Persian, Urdu, Pashto, Kurdish, Azeri Turkish, Sindhi, and Uyghur to Sorani Kurdish, which uses an alphabetic style when written in Arabic script. African Ajami languages include Hausa, Fulfulde, Mandingo, Swahili, Wolof, Kanuri, and Tamazight, together spoken by over 200 million people. Combined with non-Arabic Arabic-script languages in Asia and the Middle East, the languages within AbjadNLP’s scope represent around 1 billion speakers worldwide, many of which are severely under-resourced in NLP.

Topics of Interest

We welcome original and unpublished contributions on (but not limited to):

  • Core Technologies: Morphological analysis, tokenisation, part-of-speech tagging, named entity recognition, parsing, sentiment and emotion analysis, language modelling, and adaptation of large language models (LLMs) for low-resource contexts.

  • Applications: Machine translation, speech recognition and synthesis, OCR and handwriting recognition for complex scripts, assistive technologies, and social media/conversational AI.

  • Resources and Tools: Corpora (monolingual, parallel, and multimodal), lexicons, orthography and script documentation, text input methods, spell-checking tools, and open access datasets for under-resourced languages.

  • Cultural & Sociolinguistic Considerations: Transliteration and orthographic variation, code-switching and dialectal variation, cultural context in NLP applications, and preservation of linguistic heritage through digital tools.

Submission Guidelines

We follow the EACL 2026 submission format and guidelines.

  • Long papers: Up to 8 pages (excluding references) presenting substantial, original, and unpublished work.

  • Short papers: Up to 4 pages (excluding references) for small focused contributions, negative results, position papers, or system demonstrations.

All submissions will undergo double-blind peer review, and must adhere to the ACL Code of Ethics. Accepted papers will be published in the official EACL 2026 workshop proceedings.

Authors describing orthographies are encouraged to address the points outlined in Hosken (2003) and consider the Unicode model presented by Constable (2002) to ensure consistency and industry applicability.

Important Dates (Following EACL 2026 schedule)

  • First Call for Papers: 15 October 2025

  • Second Call for Papers: 12 November 2025

  • Third Call for Papers: 5 December 2025

  • Direct Submission Deadline: 19 December 2025

  • Pre-reviewed (ARR) Submission Deadline: 2 January 2026

  • Notification of Acceptance: 23 January 2026

  • Camera-ready Papers Due: 3 February 2026

  • Workshop Dates: 24–29 March 2026