多语言和代码混合滥用检测中毒性信号的条件可靠性
Conditional Reliability of Toxicity Signals for Multilingual and Code-Mixed Abuse Detection
浏览论文内容
中文总结 AI 辅助
研究多语言和代码混合短文本中毒性信号的条件可靠性，提出ToxGate方法，通过信任融合头根据编码器表示对辅助信号进行条件设定，实验表明该方法在多个设置下优于普通编码器，在高风险审核切片中有显著增益。
中文摘要 AI 辅助
审核系统越来越依赖外部毒性工具，但这些工具在代码混合、音译、俚语和语言不匹配情况下不可靠。我们研究印度多语言和代码混合短文本中毒性先验的条件可靠性，英语毒性、印度语滥用和基于规则的严重性线索可能是有用证据，但仅在某些语言和滥用严重性背景下。我们提出ToxGate，一个信任融合头，在将每个辅助信号添加到预测状态之前，根据编码器表示对其进行条件设定。在三个短文本滥用数据集、四个变压器编码器以及每个设置五个种子的实验中，ToxGate在12个域内设置中的10个以及8个转移设置中的7个方面优于匹配的普通编码器。最大且最具可解释性的增益出现在高风险审核切片中。更广泛的教训是，审核系统应将外部毒性工具和先验视为条件证据而非固定特征或基本事实，在聚焦消融中，特定源的门控在转移、严重滥用切片和高风险分类中给出最强结果。
英文摘要
Moderation systems increasingly rely on external toxicity tools, but those tools are unreliable under code-mixing, transliteration, slang, and language mismatch. We study the \emph{conditional reliability} of toxicity priors in Indian multilingual and code-mixed short text: English toxicity, Indic abuse, and rule-based severity cues can be useful evidence, but only in some linguistic and abuse-severity contexts. We propose ToxGate, a trust-fusion head that conditions each auxiliary signal on the encoder representation before adding it to the prediction state. Across three short-text abuse datasets, four transformer encoders, and five seeds per setting, ToxGate improves over matched plain encoders in 10 of 12 in-domain settings and 7 of 8 transfer settings. The largest and most interpretable gains occur in high-risk moderation slices, including explicit slurs, violent threats, and cross-dataset transfer. The broader lesson is that moderation systems should treat external toxicity tools and priors as conditional evidence rather than fixed features or ground truth, in focused ablations, source-specific gating gives the strongest results in transfer, severe-abuse slices, and high-risk triage.
发表机构
- Centre for Development of Advanced Computing（先进计算发展中心）
机构由 AI 辅助整理，请以论文原文为准。
