用分层软标签模型提升表情包中性别歧视识别准确率
AI Wizards at EXIST 2026: Hierarchical Soft-Label Learning for Multimodal Sexism Identification in Memes
- 将标注者分布建模为条件软标签，分层处理多任务识别
- 在任务2.3上排名第一，其他两任务位列前四
- 适合关注多模态偏见检测与轻量级模型设计的研究者
我们提交AI Wizards团队参加EXIST 2026的表情包多模态性别歧视识别任务。该任务包含三个难度递增的子任务。我们将其建模为基于经验标注者分布的条件软标签预测，采用固定Gemini Embedding 2视觉-语言表征，通过轻量级门控MLP进行训练，优化目标为KL散度，并引入同方差不确定性加权。在官方软-软排行榜上，我们的系统在任务2.3中排名第一，在任务2.1和2.2中分别位列第四。代码已开源：https://github.com/NLP-AI-Wizards/EXIST-2026。
原文摘要 · Abstract (English)
We present the AI Wizards submission to EXIST 2026 for multimodal sexism identification in memes. The task is composed of three, increasingly harder subtasks. We model them hierarchically as conditional soft-label prediction over empirical annotator distributions. Our system maps fixed Gemini Embedding 2 vision-language representations through a lightweight Gated MLP trained with KL divergence and homoscedastic uncertainty weighting. Our submissions ranked first on Task 2.3 and fourth on Tasks 2.1 and 2.2 on the official Soft-Soft leaderboards. The code is available at https://github.com/NLP-AI-Wizards/EXIST-2026
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。