偏见超越国界：多语言大语言模型中的政治意识形态评估与引导
Bias Beyond Borders: Political Ideology Evaluation and Steering in Multilingual LLMs
AI总结：
本文提出跨语言对齐引导框架，用于评估和减轻多语言LLM中的政治偏见，通过跨语言对齐意识形态表示以实现公平性与多样性的平衡。
摘要：
大型语言模型（LLMs）日益塑造全球话语，使公平性和意识形态中立性成为负责任的人工智能部署的关键。尽管对LLMs中政治偏见的关注不断增长，但先前的工作大多集中在高资源、西方语言或狭窄的多语言设置上，留下了跨语言一致性和安全的后续缓解措施的探索不足。为了解决这一差距，我们提出了一个大规模的多语言偏见评估，涵盖50个国家和33种语言。我们引入了一个互补的后续缓解框架，即跨语言对齐引导（CLAS），旨在通过在不同语言中对齐意识形态表示并动态调节干预强度来增强现有引导方法。该方法将由政治提示诱导的潜在意识形态表示对齐到一个共享的意识形态子空间，确保跨语言一致性，适应机制防止过度校正并保持一致性。实验表明，沿经济和社会轴线显著减少了偏见，同时对响应质量的损害极小。所提出的框架建立了可扩展且可解释的公平意识多语言LLM治理范式，平衡了意识形态中立性与语言和文化多样性。
英文摘要：
Large Language Models (LLMs) increasingly shape global discourse, making fairness and ideological neutrality essential for responsible AI deployment. Despite growing attention to political bias in LLMs, prior work largely focuses on high-resource, Western languages or narrow multilingual settings, leaving cross-lingual consistency and safe post-hoc mitigation underexplored. To address this gap, we present a large-scale multilingual evaluation of political bias spanning 50 countries and 33 languages. We introduce a complementary post-hoc mitigation framework, Cross-Lingual Alignment Steering (CLAS), designed to augment existing steering methods by aligning ideological representations across languages and dynamically regulating intervention strength. This method aligns latent ideological representations induced by political prompts into a shared ideological subspace, ensuring cross lingual consistency, with the adaptive mechanism prevents over correction and preserves coherence. Experiments demonstrate substantial bias reduction along both economic and social axes with minimal degradation in response quality. The proposed framework establishes a scalable and interpretable paradigm for fairness-aware multilingual LLM governance, balancing ideological neutrality with linguistic and cultural diversity.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
