Join the 2nd Arabic Named Entity Recognition Shared Task – Deadline Extended to 10 April!
The Arabic NLP community is excited to announce the 2nd Arabic Named Entity Recognition Shared Task, part of the ArabicNLP’24 conference. This year’s shared task aims to advance the state-of-the-art in Arabic NER by focusing on fine-grained entity recognition and nested entities. Participants can choose from three subtasks, each with its unique challenges and goals.
The dataset for this shared task is the new Wojood-Fine dataset, which includes subtypes of entity types for Arabic Fine-Grained Entity Recognition. The subtasks are:
- Subtask 1: Closed-Track Flat Fine-Grain NER
- Subtask 2: Closed-Track Nested Fine-Grain NER
- Subtask 3: Open-Track NER – Gaza War
For Subtask 1 and 2, participants will be provided with Wojood-Fine Flat and Nested train (70%) and development (10%) datasets. The final evaluation will be on the test set (20%). External data is not allowed for these subtasks. Subtask 3 focuses on real-world events and allows participants to use external resources and generative models in different ways, such as fine-tuned, zero-shot learning, or in-context learning.
Two baseline models trained on WojoodFine (flat and nested) are provided for participants. The code used to produce these baselines is available on GitHub. Google Colab notebooks are also available to help participants experiment with the baseline models.
To participate, interested teams must register via the NERSharedTask 2024 form by 10 April. Participating teams will be provided with common training development datasets, and each team is allowed a maximum of 3 submissions. The blind test dataset will be used to evaluate the output of the participating teams.
For any questions related to this task, please contact the organizers directly at NERSharedtask@gmail.com.
Join the Arabic NLP community in advancing Arabic NER and information extraction by participating in the 2nd Arabic Named Entity Recognition Shared Task!