用反事实校准减少泰语政治立场判断中的偏见
Debiasing Large Language Models in Thai Political Stance Detection via Counterfactual Calibration
- 通过反事实数据增强和理由监督解耦情感与立场
- 显著降低虚假关联，提升零样本泛化能力
- 适合关注低资源语言公平性的研究者
在资源匮乏且文化复杂的泰国政治语境中，大语言模型面临严峻挑战。由于语言间接、人物对立、情感与立场交织，现有模型常出现情感泄露和对特定实体的偏好等系统性偏差，影响公平性与可靠性。本文提出ThaiFACTUAL——一种轻量级、模型无关的校准框架，无需微调即可缓解政治偏见。该方法结合反事实数据增强与基于理由的监督，有效解耦情感与立场。同时发布首个高质量泰语政治立场数据集，涵盖立场、情感、推理依据及偏见标记，覆盖多样主体与事件。实验表明，ThaiFACTUAL显著降低虚假相关性，提升多模型零样本泛化性能与公平性。本工作强调了针对非主流语言的文化适配去偏技术的重要性。
原文摘要 · Abstract (English)
Political stance detection in low-resource and culturally complex settings poses a critical challenge for large language models (LLMs). In the Thai political landscape - marked by indirect language, polarized figures, and entangled sentiment and stance - LLMs often display systematic biases such as sentiment leakage and favoritism toward entities. These biases undermine fairness and reliability. We present ThaiFACTUAL, a lightweight, model-agnostic calibration framework that mitigates political bias without requiring fine-tuning. ThaiFACTUAL uses counterfactual data augmentation and rationale-based supervision to disentangle sentiment from stance and reduce bias. We also release the first high-quality Thai political stance dataset, annotated with stance, sentiment, rationales, and bias markers across diverse entities and events. Experimental results show that ThaiFACTUAL significantly reduces spurious correlations, enhances zero-shot generalization, and improves fairness across multiple LLMs. This work highlights the importance of culturally grounded debiasing techniques for underrepresented languages.
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。