StanceNakba Shared Task: Actor and Topic-Aware Stance Detection in Public Discourse
Researchers have introduced the StanceNakba 2026 shared task, focusing on stance detection within the polarized discourse surrounding the Palestinian-Israeli conflict. The task includes two subtasks: classifying English social media posts into Pro-Palestine, Pro-Israel, or Neutral stances, and identifying stances on normalization with Israel and refugee presence in Arabic posts. The initiative is supported by a dataset of 2,606 annotated posts, and participating teams utilized fine-tuned transformer models like MARBERT and AraBERT, achieving high F1 scores. AI
IMPACT Advances NLP techniques for analyzing polarized discourse in conflict zones, potentially improving understanding of social media narratives.