PropaInsight:从技巧、诉求和意图角度更深入地理解宣传
PropaInsight: Toward Deeper Understanding of Propaganda in Terms of Techniques, Appeals, and Intent
AI总结:
本文提出 propainsight 框架,从技巧、诉求和意图解析宣传,并构建融合人工与合成数据的 propagaze 数据集,显著提升 LLM 的宣传分析与跨域泛化性能。
摘要:
宣传在塑造公众舆论和助长虚假信息方面发挥着关键作用。现有研究主要关注识别宣传技巧,却缺乏捕捉此类内容更广泛动机和影响的能力。为应对这些挑战,我们提出 propainsight,这是一个以基础社会科学研究为依据的概念框架,系统地将宣传拆解为技巧、唤醒诉求和潜在意图。propainsight 为理解宣传如何在不同语境中运作提供了更细粒度的视角。此外,我们提出 propagaze,这是一个新颖数据集,将人工标注数据与通过精心设计的流程生成的高质量合成数据结合起来。实验表明,现成的 LLM 在宣传分析方面表现困难,而使用 propagaze 训练可显著提升性能。与 1-shot GPT-4-Turbo 相比,经过微调的 Llama-7B-Chat 在技巧识别中的文本跨度 IoU 提高 203.4%,在诉求分析中的 BertScore 提高 66.2%。此外,在数据稀疏和跨域场景中,propagaze 可补充有限的人工标注数据,展现出实现全面且可泛化宣传分析的潜力。
英文摘要:
Propaganda plays a critical role in shaping public opinion and fueling disinformation. While existing research primarily focuses on identifying propaganda techniques, it lacks the ability to capture the broader motives and the impacts of such content. To address these challenges, we introduce propainsight, a conceptual framework grounded in foundational social science research, which systematically dissects propaganda into techniques, arousal appeals, and underlying intent. propainsight offers a more granular understanding of how propaganda operates across different contexts. Additionally, we present propagaze, a novel dataset that combines human-annotated data with high-quality synthetic data generated through a meticulously designed pipeline. Our experiments show that off-the-shelf LLMs struggle with propaganda analysis, but training with propagaze significantly improves performance. Fine-tuned Llama-7B-Chat achieves 203.4% higher text span IoU in technique identification and 66.2% higher BertScore in appeal analysis compared to 1-shot GPT-4-Turbo. Moreover, propagaze complements limited human-annotated data in data-sparse and cross-domain scenarios, showing its potential for comprehensive and generalizable propaganda analysis.
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。