Overview

The scientific literature now grows faster than anyone can read it. New publications arrive constantly, and the claims, evidence, named entities, relations, and citations that researchers rely on remain in unstructured text rather than in any form a machine can use directly. Turning that text into structured, machine-readable knowledge is a central problem in Natural Language Processing (NLP), and many of its questions are still open. How do we reliably identify entities, concepts, and citation contexts in scientific writing? How do we recover the argument and discourse structure of a paper, judge its quality and reproducibility, and bring together findings across many documents into knowledge that researchers can query and trust?

The Fourth Workshop on Artificial Intelligence for Scientific Publications (WASP) at IJCNLP-AACL 2026 brings together researchers working to extract, structure, and mine the knowledge contained in scholarly publications. It continues a series that began as the Workshop on Information Extraction from Scientific Publications (WIESP) at AACL-IJCNLP 2022 and IJCNLP-AACL 2023, and grew into WASP at IJCNLP-AACL 2025. As large language models (LLMs) reshape how scientific text is parsed, mined, and synthesized, this iteration keeps these information extraction questions at its center while broadening to the full pipeline that turns publications into usable knowledge, from summarization and retrieval to evaluation and the responsible use of LLMs in scientific publishing. The fourth iteration of WASP will concentrate on the following topics related to AI research for and with scientific publications, but are not limited to:

  • Scientific document parsing and structured information extraction; citation context/span extraction and citation-based knowledge mining
  • Scientific named-entity recognition and concept identification
  • Scientific literature quality assessment and automated peer review assistance
  • Argument extraction and scientific discourse analysis
  • Scientific article summarisation and headline generation
  • Question-answering and fact retrieval from scientific literature
  • Pretraining and fine-tuning LLMs on scientific corpora
  • Evaluation and alignment of LLMs for scientific understanding
  • AI-assisted scientific discovery and hypothesis generation
  • Cross-lingual scientific knowledge transfer and multilingual publication processing
  • Scientific trend analysis and research gap identification
  • Reproducibility and replicability analysis in scientific publications
  • Ethical and responsible use of LLMs in scientific publishing
  • Automated metadata generation and semantic tagging of scientific publications
  • Scientific plagiarism detection and research integrity monitoring
  • Scientific collaboration network analysis and co-authorship prediction
  • Automated scientific writing assistance and manuscript generation
  • Scientific image and figure analysis in publications

We especially welcome participation from academic and research institutions, government and industry labs, publishers, and information service providers. Projects and organizations applying NLP and ML techniques in their text mining and enrichment efforts are also welcome to participate. We strongly encourage the participation of students, researchers, and science practitioners from diverse backgrounds.

Shared Task

In addition to papers, WASP will also host the AstroCLIMB shared task. Shared task authors will be invited to write their system descriptions, which will then undergo light peer review.

The shared task papers will be published with the Fourth WASP proceedings in the ACL Anthology.

Paper Submission Information

Submission will be via OpenReview. Submissions should follow the host venue formatting guidelines and template files. Submissions (Long and Short Papers) will be subject to a double-blind peer-review process. We follow the same policies as AACL 2026 regarding anonymity, preprints, and double submissions.

Submission Link: OpenReview

Submission Types

We invite papers of the following categories:

Long papers must describe substantial, original, completed, and unpublished work. Wherever appropriate, concrete evaluation and analysis should be included. Papers must not exceed eight (8) pages of content, plus unlimited pages of references. The final versions of long papers will be given one additional page of content (up to 9 pages) to accommodate reviewers’ comments.

Short papers must describe original and unpublished work. Please note that a short paper is not a shortened long paper. Instead, short papers should make a point that can be conveyed in a few pages, such as a small focused contribution, a negative result, or an interesting application nugget. Short papers must not exceed four (4) pages, plus unlimited pages of references. The final versions of short papers will be given one additional page of content (up to 5 pages) to accommodate reviewers’ comments.

We particularly invite papers, both long and short, that bring large language models to bear on extracting and structuring scientific knowledge. Because an extracted entity, relation, or summary is only useful if researchers can trust that it reflects the original paper, we are especially interested in work on reliability: faithful information extraction, hallucination detection and mitigation, uncertainty estimation, and rigorous evaluation against the literature.

All accepted papers will be published in the WASP 2026 proceedings and indexed in the ACL Anthology.

Important Dates

  • First Call for Papers: June 29, 2026
  • Second Call for Papers: July 13, 2026
  • AstroCLIMB Shared Task Registration Start + Data Release: July 27, 2026
  • AstroCLIMB Shared Task Registration deadline: September 13, 2026
  • Third Call for Papers: August 8, 2026
  • Paper submission deadline (WASP): September 14, 2026
  • System Run and Output Submission (AstroCLIMB Shared Task): September 13, 2026
  • System Paper Submission (AstroCLIMB Shared Task): September 14, 2026
  • Notification of paper acceptance (WASP + Shared Task): October 1, 2026
  • Camera-ready submission deadline (WASP + Shared Task): October 12, 2026
  • Workshop: November 9-10, 2026

All submission deadlines are 11.59 pm UTC -12 h (“Anywhere on Earth”)

Contact

For enquiries, contact Atilla Kaan Alkan at atilla.alkan@cfa.harvard.edu.