From Specialization to Generalization: Instruction-tuned LLMs for Robust Harmful Content Mitigation

2026-08-26

AI Summary

本文通过指令调优方法改进大型语言模型，专门针对有害内容（如仇恨言论）的检测和缓解，展示了在领域内及跨领域、跨语言方面的显著性能提升。

Abstract

Large language models (LLMs) demonstrate impressive performance across a wide range of general NLP tasks; however, their effectiveness in sensitive domains, such as hate speech detection, remains less clear. Prior studies comparing prompted LLMs with state-of-the-art encoder-based models (e.g., BERT variants (Roy et al., 2023; Dönmez et al., 2024)) have shown only marginal gains, suggesting that LLMs may not excel in hate speech detection or mitigation. In this work, we revisit this question through the lens of instruction tuning. By thoroughly unifying 36 English hate speech datasets spanning multiple labeling schemes, we fine-tune a generalist LLM, based on Qwen3 (Qwen Team, 2025), specifically for hate speech mitigation. Our results demonstrate not only state-of-the-art performance on in-domain benchmarks but also substantial improvements in cross-domain and cross-lingual generalization--areas where encoder-based specialist classifiers often struggle.

Problem

• Large language models
• hate speech detection
• instruction tuning
• generalization

Innovation

• instruction tuning
• hate speech mitigation
• cross-domain generalization
• cross-lingual generalization