Author Guidelines
The Journal of Niche NLP welcomes original research articles, position papers and short commentaries addressing topics beyond the mainstream of Natural Language Processing (NLP).
We encourage submissions from researchers worldwide, regardless of institutional affiliation. All manuscripts must be written in English.
Please read the following instructions carefully before submitting your manuscript.
1. Scope and Types of Contributions
Submissions must fall within the aims and scope of the Journal of Niche NLP. We particularly welcome research on underrepresented languages, historical data, interdisciplinary applications, overlooked research problems, alternative methodologies and critical perspectives on established NLP practices.
The journal accepts three types of contributions:
Research Articles
Original research presenting new findings, methods, resources, analyses or applications relevant to the journal's scope.
Position Papers
Insightful and well-argued contributions that challenge prevailing assumptions, critically examine established practices or propose alternative directions for NLP research.
Short Commentaries
Concise discussions, observations, or critiques addressing specific developments, challenges or opportunities within NLP. Short commentaries must not exceed 4 pages, excluding references.
There is no page limit for research articles or position papers.
Submissions that do not align with the journal's scope may be rejected by the editors without external peer review.
2. Manuscript Formatting
All manuscripts must use the official Journal of Niche NLP LaTeX template, available at:
Journal of Niche NLP — Official LaTeX Template
The template provides the required document structure, typography, citation style and formatting.
Manuscripts prepared using any other template or format will be rejected.
Authors must not modify the template to circumvent the journal's formatting requirements.
Manuscript Length and File Size
-
Research articles: No page limit.
-
Position papers: No page limit.
-
Short commentaries: Maximum 4 pages, excluding references.
-
Maximum PDF file size: 2 MB, regardless of article type.
Authors should ensure that figures and other graphical elements are appropriately optimized to meet the file size requirement.
Abstract
All submissions must include an abstract of approximately 100 words.
The abstract should clearly communicate the main topic, motivation and contribution of the manuscript.
3. Double-Blind Peer Review
The Journal of Niche NLP uses a double-blind peer-review process. Authors must prepare their submissions so that their identities cannot be readily determined from the manuscript.
When preparing your manuscript, enable the review mode of the official LaTeX template:
\documentclass[review]{journalofnichenlp}
This mode automatically hides author and affiliation information in the generated manuscript.
However, authors are responsible for ensuring that the rest of their submission is appropriately anonymized.
In particular:
-
Cite your own work in the third person. For example, write "Smith (2023) proposed..." rather than "In our previous work (Smith, 2023), we proposed..."
-
Do not include acknowledgements or other statements that identify the authors.
-
Do not include links to personally identifiable resources, such as your own GitHub repositories, personal websites, or project pages that reveal your identity.
-
Check that the manuscript, including its PDF metadata, does not contain identifying information.
Authors should not omit relevant self-citations merely to preserve anonymity. Previous work by the authors should be cited whenever it is relevant, just like work by any other researcher.
4. Citations and Related Work
All relevant previous research must be appropriately acknowledged and cited.
The Journal of Niche NLP places particular emphasis on recognizing related work across different languages, research communities, disciplines and methodological traditions.
Working on a niche topic does not mean working in isolation from the broader scholarly literature.
For example, a paper presenting an NLP method for Livonian should not restrict its related work to previous studies specifically concerning Livonian. Research addressing similar challenges for other endangered or low-resource languages may be equally relevant and deserves appropriate recognition.
Likewise, work on historical language processing should engage with relevant research on comparable historical datasets, even when the languages, periods or specific computational approaches differ.
No niche is so niche that there is no related research worth acknowledging.
Authors are expected to make a genuine effort to identify and cite relevant previous work, including research published outside the most prominent NLP conferences and journals.
5. Artificial Intelligence Policy
The Journal of Niche NLP recognizes that artificial intelligence has become an integral part of modern research and academic writing.
There is no shame in using AI tools.
Authors are welcome to use AI-assisted tools to help express their ideas, improve their writing, develop code, analyze data or support other aspects of the research process.
However, the ideas, scholarly contributions, and intellectual responsibility for a manuscript must remain with its human authors.
AI may help you express your thinking, but it should not replace your thinking.
Any AI-generated or AI-assisted text included in a manuscript must accurately represent the authors' own ideas, understanding and research process. Authors are responsible for the accuracy of their manuscripts, including any content produced with the assistance of AI.
In general, the use of AI tools does not need to be documented any more extensively than the use of other software tools employed in conducting research.
As with any research software, tools should be described when their use is relevant to understanding the methodology or reproducing the results.
The journal does not require special declarations merely because AI was used to assist with writing or language editing.
AI systems cannot be listed as authors. Responsibility for the manuscript rests entirely with its human authors.
6. Authorship
All human contributors who have participated in developing the research ideas, conducting any part of the research, or writing the manuscript must be appropriately included as authors.
Authorship should reflect actual intellectual and research contributions rather than academic status, institutional position or seniority.
The journal expects that all individuals who have contributed to the intellectual development, execution or writing of the research receive appropriate recognition.
All listed authors should be aware of the submission and accept responsibility for their contributions.
7. Originality and Previous Publication
All submitted manuscripts must present original, previously unpublished work.
Manuscripts that have already been formally published in a journal, conference proceedings or another scholarly publication are not eligible for submission as original contributions.
Preprints are permitted.
Authors may submit manuscripts that have previously been made publicly available as preprints, including on institutional repositories or preprint servers.
The existence of a preprint does not exempt authors from preparing an appropriately anonymized manuscript for double-blind peer review.
8. Open Access and Copyright
The Journal of Niche NLP is a fully open-access journal.
There are no submission fees, article processing charges, or publication fees.
Authors retain copyright to their published work.
All accepted articles are published under the Creative Commons Attribution (CC BY) license, which permits sharing, redistribution, and adaptation, provided that appropriate credit is given to the original authors.
9. Submission and Publication
The Journal of Niche NLP operates on a continuous submission basis.
There are no fixed submission deadlines for regular articles. Authors may submit their manuscripts at any time throughout the year.
Submissions undergo an initial editorial assessment to determine whether they meet the journal's scope and formal requirements.
Manuscripts that pass this assessment proceed to double-blind peer review.
Accepted articles are published individually as soon as practicable following acceptance and completion of the publication process, rather than being held for a scheduled issue.
We look forward to receiving your work and helping bring valuable research beyond the mainstream of NLP to a wider audience.