Debiasing Large Language Models in Thai Political Stance Detection via Counterfactual Calibration

AbstractPolitical stance detection in low-resource and culturally complex settings poses a critical challenge for large language models (LLMs). In the Thai political landscape—rich with indirect expressions, polarized figures, and sentiment-stance entanglement—LLMs often exhibit systematic biases, including sentiment leakage and entity favoritism. These biases not only compromise model fairness but also degrade predictive reliability in real-world applications. We introduce ThaiFACTUAL, a lightweight, model-agnostic calibration framework that mitigates political bias without fine-tuning LLMs. ThaiFACTUAL combines counterfactual data augmentation with rationale-based supervision to disentangle sentiment from stance and neutralize political preferences. We curate and release the first high-quality Thai political stance dataset with stance, sentiment, rationale, and bias markers across diverse political entities and events. Our results show that ThaiFACTUAL substantially reduces spurious correlations, improves zero-shot generalization, and enhances fairness across multiple LLMs. This work underscores the need for culturally grounded bias mitigation and offers a scalable blueprint for debiasing LLMs in politically sensitive, underrepresented languages.