[Due to many requests, we have extended the deadlines.]

We invite proposals for tasks to be run as part of RANLP 2025 (Recent Advances 
in Natural Language Processing): 
https://ranlp.org/ranlp2025/.<https://ranlp.org/ranlp2025/>

RANLP is one of the most influential and competitive NLP conferences. RANLP 
2025 will take place in September 2025 at the Black Sea city of Varna. For the 
first time in RANLP history, we are organising a shared task campaign as part 
of the main conference and inviting task organisers to submit their task 
proposals. Researchers and practitioners from all areas of Natural Language 
Processing and related communities are invited to submit task proposals.

For RANLP 2025, we welcome any task that can evaluate an automatic system for 
natural language processing. We especially encourage tasks for languages other 
than English, multi-lingual tasks, and tasks that develop novel applications of 
natural language processing.
We strongly encourage proposals based on already published datasets, as this 
can provide concrete examples and help minimise the challenges of organising 
the shared task. In the event of receiving many proposals, preference will be 
given to proposals based on already published datasets.

If you are unsure whether a task is suitable, please contact the shared task 
chairs to discuss your idea.


Task Selection
Task proposals will be reviewed by at least two reviewers, and the reviews will 
serve as the basis for acceptance decisions. Task proposals will be evaluated 
on:

  *
Novelty - Is the task based on a new problem that has not been explored much in 
the community? If similar tasks have been organised before, does this task 
cover new languages/ domains?
  *   Data – Is the data available and published already? Do annotations have 
meaningfully high inter-annotator agreements? Have all appropriate licenses for 
the use and re-use of the data been secured?
  *   Evaluation—Is the evaluation methodology sound? Is there an automated 
platform for the evaluation (e.g., CodaLab, Kaggle)?

Task Organisation
We specifically welcome task proposals from early career researchers. However, 
we strongly encourage tasks that have a diverse team of organisers as that will 
ease the task organisation. Apart from providing a dataset, task organisers are 
expected to:

  1.  Verify data quality in terms of annotator agreement.
  2.  Verify licenses for the data to allow its use in the competition.
  3.  Provide task participants with baseline systems.
  4.  Create a CodaLab or other similar evaluation platform for the task and 
manage automatic evaluation.
  5.  Promote the task within the target research community.
  6.  Manage and organise review process of participants’ submissions of system 
description papers.
  7.  Write a task description paper to be included in RANLP proceedings.
  8.  Contribute to the tasks overview paper written by shared task chairs and 
other task organisers which will also be included in RANLP proceedings.
  9.  Register and present the shared task description paper at RANLP 2025 on 
either 11th or 12th September 2025 (the exact date will be confirmed later)

Important Dates

  *
Task proposals due - November 20, 2024
  *
Task selection notification – November 25, 2024

Recommended Timeline for the Tasks

  *
Sample data and task website ready - December 1, 2024
  *   Training data ready - December 15, 2024
  *   Evaluation data ready - March 1, 2025
  *   Evaluation starts – March 10, 2025
  *   Evaluation end - March 31, 2025 (latest date; task organisers may choose 
an earlier date)
  *   Paper submission due – April 20, 2025
  *   Notification to authors – May 16, 2025
  *   Task overview paper due – May 25, 2025
  *   Camera-ready due - May 31, 2025
  *   Shared task presentation co-located with RANLP 2025 – September 11 and 
September 12, 2025

Tasks that do not meet critical deadlines, such as those for launching the 
task, setting up the CodaLab website, and uploading samples, training, and 
evaluation data, may be cancelled at the discretion of the shared task chairs.
Submission Details
The task proposal should be a self-contained document of no longer than 2 pages 
(plus additional pages for references). All submissions must be in PDF format, 
following the RANLP 2023 template available at 
https://ranlp.org/ranlp2023/index.php/submissions/
Each proposal should contain the following:

  *   Overview
     *   Summary of the task – What is the goal of the task
     *   Expected number of participants and justification
  *   Data & Resources
     *   How the training/testing data will be produced. Discuss whether the 
dataset is already published
     *   Details of license, so that the data can be used by the research 
community
     *   How much data will be produced
     *   How data quality will be ensured and evaluated
     *   An example of what the data would look like
  *   Evaluation
     *   The evaluation methodology to be used, including clear evaluation 
criteria  -
     *   The evaluation platform (i.e. CodaLab, Kaggle etc.)
  *   Task organisers
     *   Names, affiliations, email addresses
     *   brief description of relevant experience or expertise

The submissions should be done via START - 
https://softconf.com/ranlp25/papers/user/scmd.cgi?scmd=submitPaperCustom&pageid=3
Proceedings
Tasks overview paper, task description papers and participant papers will be 
published as part of RANLP 2025 proceedings in ACLAnthology. Task organisers 
and participants are expected to attend RANLP 2025 on September 11 and 
September 12, 2025, and present their work in order to include it in the 
proceedings.
Shared Task Chairs
Dr Tharindu Ranasinghe, Lancaster University, UK
Dr Saad Ezzini, Lancaster University, UK
RANLP 2024 Chairs
Programme Committee Chair: Prof Dr Ruslan Mitkov, Lancaster University, UK
Organising Committee Chair: Prof Dr Galia Angelova, Bulgarian Academy of 
Sciences, Bulgaria
Best Regards

Dr Tharindu Ranasinghe | Lecturer in Security and Protection Science
School of Computing and Communications | Lancaster University
Contact me on 
Teams<https://teams.microsoft.com/l/chat/0/0?users=t.ranasin...@lancaster.ac.uk>
www.lancaster.ac.uk<https://www.lancaster.ac.uk/>

Reply via email to