语音数据能降低文本毒性检测中的群体偏见，尤其在模糊或易引发争议的样本上。
On the Role of Speech Data in Reducing Toxicity Detection Bias
- 用高质量语音标注构建多语言数据集，对比语音与文本分类器的偏见表现。
- 语音模型在涉及群体提及的模糊样本上误报率显著更低。
- 改进分类器比优化语音转写更能有效减少群体偏见，适合安全敏感场景研究者。
文本毒性检测系统存在显著偏见，对提及特定人口群体的样本产生过高误报率。但语音领域的毒性检测情况如何？为探究文本偏见是否能在语音系统中缓解，我们为多语言MuTox数据集生成了一套高质量群体标注，并据此系统比较了语音与文本分类器的表现。结果表明，推理时使用语音数据可降低对群体提及的偏见，尤其在语义模糊或易引发争议的样本上效果更明显。此外，改善分类器性能比优化语音转写流程对减少群体偏见更为有效。我们公开发布标注数据，并为未来毒性检测数据集构建提供建议。
原文摘要 · Abstract (English)
Text toxicity detection systems exhibit significant biases, producing disproportionate rates of false positives on samples mentioning demographic groups. But what about toxicity detection in speech? To investigate the extent to which text-based biases are mitigated by speech-based systems, we produce a set of high-quality group annotations for the multilingual MuTox dataset, and then leverage these annotations to systematically compare speech- and text-based toxicity classifiers. Our findings indicate that access to speech data during inference supports reduced bias against group mentions, particularly for ambiguous and disagreement-inducing samples. Our results also suggest that improving classifiers, rather than transcription pipelines, is more helpful for reducing group bias. We publicly release our annotations and provide recommendations for future toxicity dataset construction.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。