Introduction

Due to the fast growth of the number of scientific publications, keeping abreast of new findings and recognizing unsolved challenges are becoming more difficult for researchers in various fields. This problem could be more severe when there is a high demand for a specific topic such as the recent COVID-19 global pandemic. As such, it is necessary to be equipped with state-of-the-art technologies to effectively combine precious findings from diverse scientific documents into a single easily accessible resource. Due to the importance of this requirement for the scientific community, there have been some efforts to achieve this goal. Moreover, new resources and tools released for scientific document processing such as SciBERT (Beltagy, Lo, and Cohan, 2019) and SciFact (Wadden, et al. 2020) provide more research opportunities for scientific document understanding. However, despite all of the recent progress, the fragmented research focusing on different aspects for this domain necessitates a forum for researchers from different perspectives to discuss achievements, new challenges, new resource requirements, and impacts of scientific document understanding on various fields. In addition to the recent focus on scholarly text processing and document understanding in natural language processing, this workshop extends SDU to other scientific areas, including but not limited to scientific image processing, automatic programming, knowledge graph manipulation, and data management. We hope that this workshop will foster collaborations from researchers working on different scientific and AI areas for SDU. Finally, we present a shared task of acronym extraction in multiple languages (i.e., English, French, Spanish, Danish, Persian, and Vietnamese) to boost the research in this area.

Topics of Interest

SDU is a workshop to gather insights into the recent advances and remaining challenges on scientific document understanding. As this topic is inherently a multi-disciplinary subject, researchers from artificial intelligence, natural language processing, information retrieval and extraction, image processing, data mining, statistics, bio-medicine, cybersecurity, finance and other fields are invited to submit papers on the recent advances, resources, tools, and upcoming challenges for SDU. Topics of interest for this workshop include but are not limited to:

Information extraction and information retrieval for scientific documents;

Question answering and question generation for scholarly documents;

Word sense disambiguation, acronym identification and expansion, and definition extraction;

Document summarization, text mining, document topic classification, and machine reading comprehension for scientific documents;

Graph analysis applications including knowledge graph construction and representation, graph reasoning and query knowledge graphs;

Multi-modal and multi-lingual scholarly text processing;

Biomedical image processing, scientific image plagiarism detection and data visualization;

Code/Pseudo-code generation from text and image/diagram captioning;

New language understanding resources such as new syntactic/semantic parsers, language models or techniques to encode scholarly text;

Survey or analysis papers on scientific document understanding and new tasks and challenges related to each scientific domain;

Factuality, data verification and anti-science detection;

Important Dates
Paper submission deadline: November 15, 2021

Notification date: December 5, 2021

Camera-ready submissions due: December 25, 2021

SDU workshop at AAAI 2022: February 28 - March 1, 2022

All deadlines are “anywhere on earth” (UTC-12)

Submission
Authors are invited to submit their unpublished work that represents novel research. The papers should be written in English using the AAAI-22 author kit and follow the AAAI 2022 formatting guidelines. Authors can also submit supplementary materials, including technical appendices, source codes, datasets, and multimedia appendices. All submissions, including the main paper and its supplementary materials, should be fully anonymized. For more information on formatting and anonymity guidelines, please refer to AAAI 2022 call for paper page.

All papers will be double-blind peer-reviewed. SDU accepts papers in two tracks:

Short Paper Track: Up to 4 pages of content including the references. Upon acceptance, the authors are provided with 1 more page to address the reviewer comments.

Long Paper Track: Up to 8 pages of content including the references. Upon acceptance, the authors are provided with 1 more page to address the reviewer comments.

Two reviewers with the same technical expertise will review each paper. Authors of the accepted papers will present their work in either the Oral or Poster session. All accepted papers will appear on the workshop proceedings that will be published on CEUR-WS. The authors will keep the copyright of their papers that are published on CEUR-WS. The workshop proceedings will be indexed by DBLP.

At least one author of each accepted paper should register at the conference and present the work at the workshop.

Submission should be done electronically in PDF format via EasyChair. SDU will not accept any submission from other mechanisms such as Email. For information on System Paper submission for the share tasks, please refer to our Shared Task page.



References

Iz Beltagy, Kyle Lo, and Arman Cohan. 2019. SciBERT: A Pretrained Language Model for Scientific Text. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP).

David Wadden, Shanchuan Lin, Kyle Lo, Lucy Lu Wang, Madeleine van Zuylen, Arman Cohan, and Hannaneh Hajishirzi 2020. Fact or Fiction: Verify-ing Scientific Claims. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP).