HateRephrase: Zero- and Few-Shot Reduction of Hate Intensity in Online Posts using Large Language Models
作者: Vibhor Agarwal, Yu Chen, Nishanth Sastry
分类: cs.CL
发布日期: 2023-10-21
💡 一句话要点
提出HateRephrase以减少在线帖子中的仇恨言论强度
🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)
关键词: 仇恨言论 大型语言模型 文本生成 自然语言处理 社交媒体 内容审核
📋 核心要点
- 现有的仇恨言论检测和反驳方法无法完全消除其社会危害，且仇恨言论传播速度快于反驳内容。
- 本文提出在帖子发布前对潜在仇恨言论进行重新措辞的策略，利用大型语言模型进行有效处理。
- 实验结果显示，GPT-3.5在生成可接受的重新措辞文本方面表现优于所有基线和开源模型，且人类评估结果也支持这一发现。
📝 摘要（中文）
仇恨言论在数字时代愈发普遍，尽管已有研究致力于检测和生成反驳言论，但这些方法无法完全消除仇恨言论带来的社会危害。本文提出了一种新颖且有效的方法，通过在帖子发布前建议重新措辞潜在的仇恨言论，利用大型语言模型（LLMs）进行实验，显示出优于现有基线模型的效果。研究开发了四种不同的提示，基于任务描述、仇恨定义、少量示例和思维链进行全面实验，评估生成文本的有效性，确保其仇恨强度降低且语义保持一致。结果表明，GPT-3.5在所有提示类型中表现最佳，甚至超越了人类生成的标准答案。
🔬 方法详解
问题定义：本文旨在解决在线帖子中仇恨言论的强度问题，现有方法在检测和反驳方面存在局限，无法有效减少仇恨言论的传播。
核心思路：提出在帖子发布前对潜在仇恨言论进行重新措辞，利用大型语言模型的生成能力，旨在降低仇恨强度而不改变原意。
技术框架：整体流程包括四个主要模块：任务描述提示、仇恨定义提示、少量示例提示和思维链提示，分别用于引导模型生成合适的文本。
关键创新：本研究的创新在于利用少量示例提示优化生成效果，且通过多种评估指标确保生成文本的有效性，显著优于现有的BART-Detox等基线模型。
关键设计：在实验中，采用了多种评估指标来衡量生成文本的仇恨强度和语义一致性，特别关注少量示例的使用效果，确保生成文本在语义上与原文相似。
🖼️ 关键图片
📊 实验亮点
实验结果表明，GPT-3.5在所有提示类型中均优于基线和开源模型，尤其在生成的重新措辞文本中，其表现甚至超过了人类生成的标准答案，显示出显著的性能提升。
🎯 应用场景
该研究的潜在应用场景包括社交媒体平台、在线评论系统和论坛等，能够有效减少仇恨言论的传播，提升网络环境的安全性和友好度。未来，该方法可扩展至其他类型的有害内容处理，具有重要的社会价值和影响力。
📄 摘要（原文）
Hate speech has become pervasive in today's digital age. Although there has been considerable research to detect hate speech or generate counter speech to combat hateful views, these approaches still cannot completely eliminate the potential harmful societal consequences of hate speech -- hate speech, even when detected, can often not be taken down or is often not taken down enough; and hate speech unfortunately spreads quickly, often much faster than any generated counter speech. This paper investigates a relatively new yet simple and effective approach of suggesting a rephrasing of potential hate speech content even before the post is made. We show that Large Language Models (LLMs) perform well on this task, outperforming state-of-the-art baselines such as BART-Detox. We develop 4 different prompts based on task description, hate definition, few-shot demonstrations and chain-of-thoughts for comprehensive experiments and conduct experiments on open-source LLMs such as LLaMA-1, LLaMA-2 chat, Vicuna as well as OpenAI's GPT-3.5. We propose various evaluation metrics to measure the efficacy of the generated text and ensure the generated text has reduced hate intensity without drastically changing the semantic meaning of the original text. We find that LLMs with a few-shot demonstrations prompt work the best in generating acceptable hate-rephrased text with semantic meaning similar to the original text. Overall, we find that GPT-3.5 outperforms the baseline and open-source models for all the different kinds of prompts. We also perform human evaluations and interestingly, find that the rephrasings generated by GPT-3.5 outperform even the human-generated ground-truth rephrasings in the dataset. We also conduct detailed ablation studies to investigate why LLMs work satisfactorily on this task and conduct a failure analysis to understand the gaps.