HARE：基于逐步推理的可解释仇恨言论检测
HARE: Explainable Hate Speech Detection with Step-by-Step Reasoning
AI总结：
本文提出仇恨言论检测框架HARE，利用大型语言模型的逐步推理能力填补现有标注方案的推理空白，从而有效监督检测模型，在SBIC和Implicit Hate基准上超越基线并提升解释质量与泛化能力。
摘要：
随着社交媒体的激增，准确检测仇恨言论已成为确保网络安全的当务之急。为了应对细微形式的仇恨言论，识别并详尽解释仇恨言论以帮助用户理解其有害影响至关重要。近期的基准测试试图通过在仇恨文本隐含意义的自由文本标注上训练生成模型来解决这一问题。然而，我们发现现有的标注方案存在显著的推理差距，这可能阻碍检测模型的监督。在本文中，我们提出了一个仇恨言论检测框架 HARE，它利用大型语言模型（LLMs）的推理能力来填补仇恨言论解释中的这些空白，从而实现对检测模型的有效监督。在 SBIC 和 Implicit Hate 基准上的实验表明，我们的方法使用模型生成的数据，在使用现有自由文本人工标注的基线上始终表现更佳。分析表明，我们的方法提升了已训练模型的解释质量，并改善了对未见数据集的泛化能力。我们的代码可在 https://github.com/joonkeekim/hare-hate-speech.git 获取。
英文摘要：
With the proliferation of social media, accurate detection of hate speech has become critical to ensure safety online. To combat nuanced forms of hate speech, it is important to identify and thoroughly explain hate speech to help users understand its harmful effects. Recent benchmarks have attempted to tackle this issue by training generative models on free-text annotations of implications in hateful text. However, we find significant reasoning gaps in the existing annotations schemes, which may hinder the supervision of detection models. In this paper, we introduce a hate speech detection framework, HARE, which harnesses the reasoning capabilities of large language models (LLMs) to fill these gaps in explanations of hate speech, thus enabling effective supervision of detection models. Experiments on SBIC and Implicit Hate benchmarks show that our method, using model-generated data, consistently outperforms baselines, using existing free-text human annotations. Analysis demonstrates that our method enhances the explanation quality of trained models and improves generalization to unseen datasets. Our code is available at https://github.com/joonkeekim/hare-hate-speech.git.
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。
