Long Papers
Must describe substantial, original, completed and unpublished work, with up to eight pages of content, plus unlimited pages for limitations, ethical considerations, and references.
NLP/CORE Unranked
ARR - August 2026
ACL Rolling Review - August 2026
Jun 23 2026Official conference site Link status unverified
ACL Rolling Review (ARR) is a centralized, rolling peer-review platform for the Association for Computational Linguistics that decouples the reviewing process from conference acceptance decisions. It allows authors to submit papers every two months, receive detailed reviews and meta-reviews, and then choose among participating ACL conferences to commit their work for acceptance, promoting efficiency, iterative improvement, and consistency in NLP research evaluation.
Key deadlines
Verified September 9, 2026Full paper
August 4, 2026
AoE
Conference timeline
Submission and decisions
Full paperKey deadline
August 4, 2026 · AoE
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Short Papers
Must describe original and unpublished work with a small, focused contribution, up to four pages of content, plus unlimited pages for limitations, ethical considerations, and references.
Approaches to low-resource settings
Submissions focusing on methods for low-resource language scenarios.
Approaches low compute settings-efficiency
Submissions focused on computational efficiency and low-resource model deployment.
Data resources
Submissions introducing new datasets or resources for NLP research.
Data analysis
Submissions presenting analytical insights from existing data.
Model analysis & interpretability
Submissions analyzing or interpreting model behavior and representations.
NLP engineering experiment
Submissions reporting on engineering experiments in NLP systems.
Publicly available software and/or pre-trained models
Submissions releasing open-source software or pre-trained models with documentation.
Position papers
Submissions presenting a viewpoint or argument on a topic in NLP.
Reproduction study
Submissions attempting to reproduce prior work and reporting results.
Surveys
Submissions that go beyond standard literature reviews, establishing new interdisciplinary connections or broad syntheses.
Theory
Submissions presenting theoretical contributions to computational linguistics or NLP.
Research areas in scope
01
Topic Areas
Clinical and Biomedical ApplicationsComputational Social Science and Cultural AnalyticsDialogue and Interactive SystemsDiscourse and PragmaticsEfficient Methods for NLPEthics, Bias, and FairnessGenerationHuman-Centered NLP and Human-AI InteractionInformation ExtractionInformation Retrieval and Text MiningInterpretability and Analysis of Models for NLPLanguage ModelingLLM AgentsLinguistic Theories, Cognitive Modeling, and PsycholinguisticsMachine Learning for NLPMachine TranslationMultilingualism and Cross-Lingual NLPMultimodality and Language Grounding to Vision, Robotics and BeyondNLP and Code ModelsNLP and Symbolic ReasoningNLP ApplicationsPhonology, Morphology, and Word SegmentationQuestion AnsweringResources and EvaluationSemantics: Lexical and Sentence-LevelSentiment Analysis, Stylistic Analysis, and Argument MiningSpeech Recognition, Text-to-Speech and Spoken Language UnderstandingSummarizationSyntax: Tagging, Chunking and Parsing
02
Theme Tracks
Theme tracks (proposed by participating publication venues and specific to those venues)
Policies worth checking twice
- All submitting authors must register as reviewers or meta-reviewers (Area Chairs) and fulfill assigned reviewing duties; failure to do so results in ineligibility to submit in the next cycle.
- Papers must be original, unpublished, and not under review elsewhere during the ARR review period; dual submission to any journal or conference is prohibited.
- Papers must be anonymized: no author names, affiliations, or self-references that reveal identity; supplementary materials must also be anonymized.
- Long papers are limited to 8 pages of content; short papers to 4 pages; both require a mandatory limitations section and may include optional ethical considerations.
- Resubmissions must link to the prior version and include a detailed explanation of revisions and point-by-point responses to previous reviews.
- Resubmissions with a previous meta-review score of 1.5 cannot be resubmitted in the next cycle; those with score of 1 can never be resubmitted.
- Authors may not change the author list after submission, except in resubmissions where new authors must be justified and removed authors must have their contributions removed.
- Authors may not submit papers that overlap significantly in content with previously published work; text reuse must not exceed 10% of total tokens.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Official conference siteCall for papersAuthors guidelinesCommon submission problemsAnnouncement for authors and reviewers submitting to January ARR December 22, 20Here
Last verified September 9, 2026