Dear colleagues,
You are invited to participate in the shared task "DAGPap24: Detecting automatically generated scientific papers" collocated with the 4th Workshop on Scholarly Document Processing (SDP 2024) to be held at ACL 2024. The competition will be held on CodaBench, launching on April 2nd, 2024. Participants are also invited to submit papers describing their findings.
Papers must follow the ACL format and conform to the ACL 2024 Submission Guidelines. Paper submission has to be done through openreview.net:
- Website: https://sdproc.org/2024/sharedtasks.html#dagpap
- Submission site: http://tinyurl.com/sdpproc
- OpenReview: https://openreview.net/group?id=aclweb.org/ACL/2024/Workshop/SDProc
- Competition launch: April 2nd – April 30th, 2024
- Paper submission deadline: May 17th, 2024
== Monetary prizes ==
We offer the following monetary prizes for the winners of the shared task:
- 1st place – $3000
- 2nd place – $1200
- 3rd place – $800
The winners will be defined based on the corresponding CodaBench competition (to be launched on April 2nd, 2024). Additional conditions are the reproducibility of the solution and a technical report submitted to the workshop.
== Call for Research Papers ==
A big problem with the ubiquity of Generative AI is that it has now become very easy to generate fake scientific papers. This can erode public trust in science and attack the foundations of science: are we standing on the shoulders of robots? The Detecting Automatically Generated Papers (DAGPap) competition aims to encourage the development of robust, reliable AI-generated scientific text detection systems, utilizing a diverse dataset and varied machine learning models in a number of scientific domains.
Building on top of a similar competition held in 2022 (https://www.kaggle.com/competitions/detecting-generated-scientific-papers), we are now looking into full texts and a more fine-grained detection of LLM-generated scientific content, when artificial content might be interspersed with human-generated one.
Participants are invited to submit papers describing their findings during the competition.
== Topics of Interest ==
We are not only interested in seeing technical reports describing competitors’ approaches to solving the Shared Task but also invite submissions on related topics:
- Detection of LLM-generated texts
- Robustness of LLM-detectors to data drift
- Specifics of LLM-detectors in the scientific domain
- Explainability of LLM detection
For an even broader range of topics, please check out the CFP for Scholarly Document Processing workshop SDP 2024 https://sdproc.org/2024/cfp.html#topics
== Submission Information ==
Authors are invited to submit full and short papers with unpublished, original work. Submissions will be subject to a double-blind peer-review process. Accepted papers will be presented by the authors at the workshop either as a talk or a poster. All accepted papers will be published in the workshop proceedings (proceedings from previous years can be found here: https://aclanthology.org/venues/sdp/).
The submissions must be in PDF format and anonymized for review. All submissions must be written in English and follow the ACL 2024 formatting requirements:
- Long paper submissions: up to 8 pages of content, plus unlimited references.
- Short paper submissions: up to 4 pages of content, plus unlimited references.
- Paper submission Website: Paper submission has to be done through https://openreview.net/group?id=aclweb.org/ACL/2024/Workshop/SDProc.
Final versions of accepted papers will be allowed 1 additional page of content so that reviewer comments can be taken into account.
== Important Dates (Shared Task) ==
- Competition launch: April 2nd (Tuesday), 2024
- Competition end: April 30th (Tuesday), 2024
- Paper submission deadline: May 17th (Friday), 2024
- Notification of acceptance: June 17th (Monday), 2024
- Camera-ready paper due: July 1st (Monday), 2024
- Workshop dates: August 15th–16th, 2024
== Organizing Committee ==
Savvas Chamezopoulos, Elsevier, the Netherlands
Yury Kashnitsky, Elsevier, the Netherlands
Anita de Waard, Elsevier, the Netherlands
Domenic Rosati, scite.ai, USA
Drahomira Herrmannova, Elsevier, USA