StanceEval 2026：第二届立场检测共享任务

arXiv cs.CL 论文
stance-detection shared-task nlp arabic-nlp cross-target-generalization llms

摘要

StanceEval 2026 是面向阿拉伯语社交媒体的第二届立场检测共享任务，重点关注跨目标（cross-target）与跨领域（cross-domain）的泛化能力。表现最佳的系统 F_avg2 分别达到 0.8994 和 0.9400，显著超越基线方法。

arXiv:2610.03215v1 Announce Type: new Abstract: StanceEval 2026 is the second edition of the StanceEval shared task series on stance detection in Arabic social media text. Stance detection aims to identify a writer's stance toward a given topic. Given a tweet and a target, participating systems must determine whether the writer's stance is Favor, Against, or None. This edition focuses on cross-target generalization across two distinct evaluation tracks: Track 1 evaluates thematically related cross-target transfer (testing on Women Driving, related to Women Empowerment from training data), while Track 2 evaluates cross-domain transfer to completely unseen targets (E-Cars and Trimester System). The shared task attracted 80 registered teams from 12 countries. During the evaluation phase, 30 unique teams submitted entries, with 21 teams officially ranked in Track 1 and 13 in Track 2 following validation filtering, and 20 teams submitting system-description papers. Participating teams employed diverse methodologies, including fine-tuned pretrained language models, prompt-based and retrieval-augmented large language models (LLMs), fine-tuned LLMs, and hybrid cascades. Top systems achieved impressive $F_{avg2}$ scores of 0.8994 on Track 1 and 0.9400 on Track 2, substantially outperforming the strongest baselines (0.7366 and 0.7475, respectively), where $F_{avg2}$ denotes the macro-averaged F1 score over the Favor and Against classes. Counterintuitively, performance on the unseen targets was higher than on the related target, a disparity could be driven by extreme target polarization, class imbalance, and dialectal or sarcastic nuance across topics.
查看原文
查看缓存全文

缓存时间: 2026/10/05 10:06 

# StanceEval 2026: The Second Stance Detection Shared Task
Source: [https://arxiv.org/abs/2610.03215](https://arxiv.org/abs/2610.03215)
[View PDF](https://arxiv.org/pdf/2610.03215)

> Abstract:StanceEval 2026 is the second edition of the StanceEval shared task series on stance detection in Arabic social media text\. Stance detection aims to identify a writer's stance toward a given topic\. Given a tweet and a target, participating systems must determine whether the writer's stance is Favor, Against, or None\. This edition focuses on cross\-target generalization across two distinct evaluation tracks: Track 1 evaluates thematically related cross\-target transfer \(testing on Women Driving, related to Women Empowerment from training data\), while Track 2 evaluates cross\-domain transfer to completely unseen targets \(E\-Cars and Trimester System\)\. The shared task attracted 80 registered teams from 12 countries\. During the evaluation phase, 30 unique teams submitted entries, with 21 teams officially ranked in Track 1 and 13 in Track 2 following validation filtering, and 20 teams submitting system\-description papers\. Participating teams employed diverse methodologies, including fine\-tuned pretrained language models, prompt\-based and retrieval\-augmented large language models \(LLMs\), fine\-tuned LLMs, and hybrid cascades\. Top systems achieved impressive $F\_\{avg2\}$ scores of 0\.8994 on Track 1 and 0\.9400 on Track 2, substantially outperforming the strongest baselines \(0\.7366 and 0\.7475, respectively\), where $F\_\{avg2\}$ denotes the macro\-averaged F1 score over the Favor and Against classes\. Counterintuitively, performance on the unseen targets was higher than on the related target, a disparity could be driven by extreme target polarization, class imbalance, and dialectal or sarcastic nuance across topics\.

## Submission history

From: Rasha Albalawi \[[view email](https://arxiv.org/show-email/33641360/2610.03215)\] **\[v1\]**Fri, 2 Oct 2026 12:32:11 UTC \(46 KB\)

相似文章