AI & ML
PersonaHate: A Scalable Persona-Based Data Synthesis Pipeline for Hate Speech Analysis
arXiv cs.CR0 views
arXiv:2611.21951v1 Announce Type: new Abstract: Hate speech remains a persistent and evolving threat on online platforms. The widespread use of large language models (LLMs) further intensifies this problem by making hateful content easier and cheaper to generate at scale. Although stronger hate speech detectors continue to be proposed, their performance remains uneven across target identity groups and different forms of hateful expression, partly because existing training data provides only limited coverage of these cases. Thus, we introduce PersonaHate, a scalable persona-driven data synthesis pipeline for constructing diverse hate speech datasets. PersonaHate begins with a large persona pool of 457K personas drawn from two real-world sources. Conditioned on selected personas from this pool, it prompts eight LLMs to generate hate speech across 34 identity groups and performs multi-judge annotation to obtain reliable labels. The resulting corpus contains 791K samples, from which we curate a 67K group-balanced training set. Our modular design also supports extension to new identity groups, generators, and emerging hate targets. Built upon this dataset, we evaluate two detector training strategies: training from scratch and further fine-tuning. Results show that all trained models achieve competitive performance across four widely used external benchmarks. Notably, a 184M-parameter DeBERTa-v3 model trained on PersonaHate reaches 77.5% average F1 and 79.4% average accuracy, outperforming OpenAI's Moderation API on HateBenchSet and MHS. Furthermore, fine-tuning Llama-3.1-8B raises its average F1 from 70.7% to 78.4%, a gain of 7.7 percentage points, surpassing LlamaGuard-3, Meta's official safety model based on the same Llama-3.1-8B architecture, on three of four benchmarks. These results show that PersonaHate can substantially improve the robustness and cross-dataset generalization of hate speech detectors.
Source:https://arxiv.org/abs/2611.21951PreviousCNN Architecture Comparison for Malware Classification with Limited Training DataNextEight in Ten LLMs Leak System-Prompt Secrets: Nevan, an Agentic Capture-the-Flag Study
Related posts
Microsoft’s new Licensed Weights lets businesses customize AI models with their own data
0ZDNet

“Collaboration over competition”: Hugging Face hands model inference traffic to OpenRouter
0The New Stack
Robots learn human movement patterns to find paths through dense crowds
0Phys.org Tech
Researchers find eight ways AI could lead to our downfall
0Phys.org Tech
Q&A: Why transportation is critical to rebuilding countries after war and terrorism
0Phys.org Tech
Wine Merges Remote OpenGL Swapchain Image Presentation, CEF Now Works With Wayland
0Phoronix