@inproceedings{chuang-etal-2024-beyond,
    title = "Beyond Demographics: Aligning Role-playing {LLM}-based Agents Using Human Belief Networks",
    author = "Chuang, Yun-Shiuan  and
      Nirunwiroj, Krirk  and
      Studdiford, Zach  and
      Goyal, Agam  and
      Frigo, Vincent V.  and
      Yang, Sijia  and
      Shah, Dhavan V.  and
      Hu, Junjie  and
      Rogers, Timothy T.",
    editor = "Al-Onaizan, Yaser  and
      Bansal, Mohit  and
      Chen, Yun-Nung",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2024",
    month = nov,
    year = "2024",
    address = "Miami, Florida, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2024.findings-emnlp.819/",
    doi = "10.18653/v1/2024.findings-emnlp.819",
    pages = "14010--14026",
    abstract = "Creating human-like large language model (LLM) agents is crucial for faithful social simulation. Having LLMs role-play based on demographic information sometimes improves human likeness but often does not. This study assessed whether LLM alignment with human behavior can be improved by integrating information from empirically-derived human belief networks. Using data from a human survey, we estimated a belief network encompassing 64 topics loading on nine non-overlapping latent factors. We then seeded LLM-based agents with an opinion on one topic, and assessed the alignment of its expressed opinions on remaining test topics with corresponding human data. Role-playing based on demographic information alone did not align LLM and human opinions, but seeding the agent with a single belief greatly improved alignment for topics related in the belief network, and not for topics outside the network. These results suggest a novel path for human-LLM belief alignment in work seeking to simulate and understand patterns of belief distributions in society."
}