arXiv cs.AI· Sergei Berezin, Reza Farahbakhsh, Noel Crespi·· 3 天前AI 评分27
毒性是交际事件的属性：面向 NLP 的语境压力测量框架（CSF）
Toxicity Is a Property of Communicative Events: A Contextual Stress Measurement Framework for NLP
AI 导读
这篇立场论文提出 Contextual Stress Framework（CSF），主张毒性检测不应被视为孤立文本的单标签分类，而应作为对情境化交际伤害的语境测量，将毒性定义为感知规范违背与所引发压力或扰乱之间的关系。CSF 解释了文本内在型检测器为何会过度标记方言或 reclaimed 语言、漏检编码或语用滥用，并在保义变换下表现脆弱。
正文 · 原文
Abstract:Toxicity detection has become core safety infrastructure for online moderation, dataset filtering, and deployed language-model systems. Yet most detectors still treat toxicity as an intrinsic property of isolated text. This position paper argues that toxicity detection should be evaluated as the contextual measurement of situated communicative harm, rather than as single-label text classification. Toxicity is not contained in words alone; it emerges when a communicative act is interpreted by an audience within a normative and social context.
We introduce the Contextual Stress Framework (CSF), which defines toxicity as a relation between perceived norm violation and induced stress or disruption. CSF explains why text-intrinsic detectors overflag dialectal or reclaimed language, miss coded or pragmatic abuse, and remain brittle under meaning-preserving transformations.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2503.16072 [cs.LG] |
| (or arXiv:2503.16072v5 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2503.16072 arXiv-issued DOI via DataCite |
Submission history
From: Sergey Berezin [view email]
[v1]
Thu, 20 Mar 2025 12:09:01 UTC (34 KB)
[v2]
Sat, 31 May 2025 12:02:49 UTC (61 KB)
[v3]
Tue, 6 Jan 2026 11:29:18 UTC (329 KB)
[v4]
Mon, 11 May 2026 16:55:53 UTC (82 KB)
[v5]
Sat, 10 Oct 2026 14:36:00 UTC (94 KB)
来源：arXiv cs.AI · arxiv.org