@inproceedings{wan-etal-2026-insideout,
    title = "{I}nside{O}ut: Measuring and Mitigating Insider{--}Outsider Bias in Interview Script Generation",
    author = "Wan, Yixin  and
      Chen, Xingrun  and
      Chang, Kai-Wei",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-nlpsi/2026.acl-long.1094/",
    doi = "10.18653/v1/2026.acl-long.1094",
    pages = "23864--23883",
    ISBN = "979-8-89176-390-6",
    abstract = "Advancements in Large language models (LLMs) have enabled a variety of downstream applications like story and interview script generation.However, recent research raised concerns about culture-related fairness issues in LLM-generated content.In this work, we identify and systematically investigate LLMs' \textbf{insider-outsider bias}, a phenomenon where models position themselves as ``insiders'' of mainstream cultures during generation while externalizing less dominant cultures.We propose the \textit{\textbf{InsideOut}} benchmark with 4,000 generation prompts and three evaluation metrics to quantify this bias through a \textit{culturally situated interview script generation} task, in which an LLM is positioned as a reporter interviewing local people across 10 diverse cultures.Empirical evaluation on 5 state-of-the-art LLMs reveals that while models adopt insider tones in over 88{\%} US-contexted scripts on average, they disproportionately default to ``outsider'' stances for non-Western cultures.To mitigate these biases, we propose \textit{2 inference-time methods}: a baseline prompt-based \textbf{Fairness Intervention Pillars (FIP)} method, and a structured \textbf{Mitigation via Fairness Agents (MFA)} framework consisting of a Single-Agent (MFA-SA), a Hierarchical-Agent (MFA-HA), and an autonomous Agentic Planning (MFA-Plan) pipeline.Empirical results demonstrate that agent-based MFA methods achieve outstanding and robust performance in mitigating the insider-outsider bias:For instance, on the Cultural Alignment Gap (CAG) metric, \textit{MFA-SA reduces bias in Llama model by 89.70 {\%} and MFA-HA mitigates bias in Qwen by 82.54{\%}}.These findings showcase the effectiveness of agent-based methods as a promising direction for mitigating biases in generative LLMs."
}