用少量样本生成的语义原型，可同时识别显性和隐性仇恨言论。
HatePrototypes: Interpretable and Transferable Representations for Implicit and Explicit Hate Speech Detection
- 基于50个样本构建类别级语义原型，实现跨任务迁移。
- 无需参数更新，仅用原型即可有效检测显性和隐性仇恨言论。
- 适合追求高效、可解释的仇恨言论检测研究者使用。
针对仇恨内容过滤模型在不同类型仇恨言论上的优化，通常依赖于在新基准上持续预训练或微调。然而现有基准主要关注对受保护群体的显性仇恨，常忽略隐性或间接仇恨，如贬低性比较、排斥或暴力呼吁、微妙歧视语言等仍会造成伤害。显性仇恨可通过表面特征捕捉，而隐性仇恨需深层全模型语义处理。本文探讨了是否需要重复微调，并分析了由语言模型优化得到的仇恨原型（HatePrototypes）的作用。我们发现，仅用每类50个样本构建的原型，即可在显性与隐性仇恨之间实现跨任务迁移，且原型可在不同基准间互换使用。此外，我们展示了基于原型的无参数早期退出机制对两类仇恨均有效。我们已开源代码、原型资源和评估脚本，以支持未来高效且可迁移的仇恨言论检测研究。
原文摘要 · Abstract (English)
Optimization of offensive content moderation models for different types of hateful messages is typically achieved through continued pre-training or fine-tuning on new hate speech benchmarks. However, existing benchmarks mainly address explicit hate toward protected groups and often overlook implicit or indirect hate, such as demeaning comparisons, calls for exclusion or violence, and subtle discriminatory language that still causes harm. While explicit hate can often be captured through surface features, implicit hate requires deeper, full-model semantic processing. In this work, we question the need for repeated fine-tuning and analyze the role of HatePrototypes, class-level vector representations derived from language models optimized for hate speech detection and safety moderation. We find that these prototypes, built from as few as 50 examples per class, enable cross-task transfer between explicit and implicit hate, with interchangeable prototypes across benchmarks. Moreover, we show that parameter-free early exiting with prototypes is effective for both hate types. We release the code, prototype resources, and evaluation scripts to support future research on efficient and transferable hate speech detection.
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。