← 研踪 PaperTrailarXiv:2510.12316 · v2
Beating Hateful Narratives on Social Media with Facts: Retrieval-Augmented Counter-Speech Generation
首次提交 2025-10-14 17:20
版本更新 2026-11-16 18:31
雷达观测 2026-11-17 17:11
摘要
Online hate speech threatens social cohesion by amplifying discrimination and harmful narratives against marginalized communities. While counter-speech (CS) - fact-based, respectful responses to hate - offers a constructive alternative to content removal, its large-scale generation remains constrained by factual reliability and the risk of generic responses. We address these challenges with a Retrieval-Augmented Generation (RAG) pipeline grounded in the first large-scale, curated knowledge base built from authoritative international and regional human rights sources, including the United Nations Digital Library, EUR-Lex, and the EU Agency for Fundamental Rights. The framework covers eight groups frequently targeted by online abuse. We systematically evaluate all pipeline components demonstrating that each design choice is empirically motivated, combining automatic metrics, LLM-based judgments, and human evaluation with comparisons against strong baselines. Results show that RAG improves factual grounding, diversity, and perceived effectiveness over No-RAG and performs favorably against existing systems. Our human evaluation further reveals differences between in-group and out-group perceptions of CS, with RAG narrowing but not eliminating this gap. These findings highlight the potential of knowledge grounded CS for scalable social media interventions, while underscoring the need to evaluate such systems with and for the communities they are intended to serve.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-16 18:31 · 最新观测版本
Beating Hateful Narratives on Social Media with Facts: Retrieval-Augmented Counter-Speech Generation
Online hate speech threatens social cohesion by amplifying discrimination and harmful narratives against marginalized communities. While counter-speech (CS) - fact-based, respectful responses to hate - offers a constructive alternative to content removal, its large-scale generation remains constrained by factual reliability and the risk of generic responses. We address these challenges with a Retrieval-Augmented Generation (RAG) pipeline grounded in the first large-scale, curated knowledge base built from authoritative international and regional human rights sources, including the United Nations Digital Library, EUR-Lex, and the EU Agency for Fundamental Rights. The framework covers eight groups frequently targeted by online abuse. We systematically evaluate all pipeline components demonstrating that each design choice is empirically motivated, combining automatic metrics, LLM-based judgments, and human evaluation with comparisons against strong baselines. Results show that RAG improves factual grounding, diversity, and perceived effectiveness over No-RAG and performs favorably against existing systems. Our human evaluation further reveals differences between in-group and out-group perceptions of CS, with RAG narrowing but not eliminating this gap. These findings highlight the potential of knowledge grounded CS for scalable social media interventions, while underscoring the need to evaluate such systems with and for the communities they are intended to serve.
查看此版本解读 →查看此版本原文 ↗