PulseAugur
中
实时 07:04:13
English(EN) SAFE-MEME: Structured Reasoning Framework for Robust Hate Speech Detection in Memes

新框架增强了meme中仇恨言论的检测能力

作者 PulseAugur 编辑部 · [1 个来源] ·

研究人员开发了SAFE-MEME，一个旨在提高meme中仇恨言论检测能力的结构化推理框架。该框架采用新颖的多模态思维链方法，结合问答式推理（SAFE-MEME-QA）和分层分类方法（SAFE-MEME-H）。该系统在两个新数据集MHS和MHS-Con上进行了基准测试，这两个数据集在常规和混淆场景中都包含细粒度的仇恨抽象。SAFE-MEME-QA在性能上优于现有的开源模型，而SAFE-MEME-H在某些指标上取得了与GPT-4o和Gemini 2.5等闭源模型相当的结果。 AI

影响 这项研究可能带来更鲁棒的内容审核AI系统，特别是在理解细微和依赖上下文的有害内容方面。

排序理由 该集群描述了一篇介绍用于特定AI任务的新框架和数据集的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

新框架增强了meme中仇恨言论的检测能力

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群描述了一篇介绍用于特定AI任务的新框架和数据集的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, safety
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
72 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.CL TIER_1 English(EN) · Palash Nandi, Shivam Sharma, Tanmoy Chakraborty ·

    SAFE-MEME：用于 memetic 内容中鲁棒仇恨言论检测的结构化推理框架

    arXiv:2412.20541v2 Announce Type: replace Abstract: Memes act as cryptic tools for sharing sensitive ideas, often requiring contextual knowledge to interpret them correctly. It makes multimodal meme moderation difficult, as existing work either lacks high-quality datasets for nua…

相关实体

相关话题