AI 巫师在 EXIST 2026：用於迷因多模态性别歧视识别之分層軟標籤學習

AI Wizards at EXIST 2026: Hierarchical Soft-Label Learning for Multimodal Sexism Identification in Memes

July 5, 2026

作者: Matteo Fasulo, Antonio Gravina, Luca Tedeschini, Luca Babboni

cs.AI

摘要

我們在EXIST 2026中提出AI Wizards團隊針對迷因多模態性別歧視識別的提交成果。該任務包含三個難度遞增的子任務。我們將其分層建模為基於經驗標註者分佈的條件式軟標籤預測。我們的系統透過輕量級閘控多層感知器（Gated MLP）處理固定的Gemini Embedding 2視覺語言表徵，並使用KL散度與同方差不確定性加權進行訓練。我們的提交在官方軟-軟排行榜上，於任務2.3獲得第一名，任務2.1與2.2獲得第四名。程式碼已公開於 https://github.com/NLP-AI-Wizards/EXIST-2026。

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