arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.15946cs.LGcs.AIcs.CR
收藏

跌入陷阱,获得智慧:通过误判风险模式检索的认知引导有害迷因检测

Fall into a Pit, Gain in a Wit: Cognitive-Guided Harmful Meme Detection via Misjudgment Risk Pattern Retrieval

State Key Laboratory of Complex System Modeling and Simulation Technology(复杂系统建模与仿真技术国家重点实验室);Science and Technology on Integrated Information System Laboratory(集成信息系统技术研究所);Institute of Software Chinese Academy of Sciences(中国科学院软件研究所);University of Chinese Academy of Sciences(中国科学院大学)

Wenshuo Wang, Ziyou Jiang, Junjie Wang, Mingyang Li, Jie Huang, Yuekai Huang, Zhiyuan Chang, Feiyan Duan, Qing Wang

AI总结:

本文提出PatMD方法,通过学习并主动缓解潜在误判风险,识别有害迷因的深层误判风险模式,提升多模态大语言模型的检测能力,实验显示在5项有害检测任务中,F1-score和准确率均显著提升。

摘要:

互联网迷因已发展为一种流行的多模态媒介,但它们正被利用微妙的修辞手法如讽刺和隐喻来传播有害观点。现有检测方法,包括基于多模态大语言模型(MLLM)的技术,在处理这些隐含表达时存在困难,导致频繁误判。本文提出PatMD,一种通过学习并主动缓解这些潜在误判风险来检测有害迷因的新方法。我们的核心思想是超越表面内容级匹配,而是识别潜在误判风险模式,主动引导MLLM避免已知的误判陷阱。我们首先构建一个知识库,其中每个迷因被分解为一个误判风险模式,解释为什么它可能被误判,无论是忽视有害含义(假阴性)还是过度解释无害内容(假阳性)。对于给定的目标迷因,PatMD检索相关模式并利用它们动态引导MLLM的推理。在包含6,626个迷因的基准测试中,PatMD在5项有害检测任务中优于最先进的基线,平均F1分数提高8.30%,准确率提高7.71%,并在未见过的和对抗性迷因上表现出一致的鲁棒性。

英文摘要:

Internet memes have emerged as a popular multimodal medium, yet they are increasingly weaponized to convey harmful opinions through subtle rhetorical devices like irony and metaphor. Existing detection approaches, including Multimodal Large Language Model (MLLM)-based techniques, struggle with these implicit expressions, leading to frequent misjudgments. This paper introduces PatMD, a novel approach that detects harmful memes by learning from and proactively mitigating these potential misjudgment risks. Our core idea is to move beyond superficial content-level matching and instead identify the underlying misjudgment risk patterns, proactively guiding the MLLMs to avoid known misjudgment pitfalls. We first construct a knowledge base where each meme is deconstructed into a misjudgment risk pattern explaining why it might be misjudged, either overlooking harmful undertones (false negative) or overinterpreting benign content (false positive). For a given target meme, PatMD retrieves relevant patterns and utilizes them to dynamically guide the MLLM's reasoning. Experiments on a benchmark of 6,626 memes across 5 harmful detection tasks show that PatMD outperforms state-of-the-art baselines, achieving an average of 8.30% improvement in F1-score and 7.71% improvement in accuracy, while exhibiting consistent robustness on unseen and adversarial memes.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑