@inproceedings{dey-etal-2026-gravity,
    title = "{GRAVITY}: A Framework for Personalized Text Generation via Profile-Grounded Synthetic Preferences",
    author = "Dey, Priyanka  and
      Rosa, Daniele  and
      Zheng, Wenqing  and
      Barcklow, Daniel  and
      Zhao, Jieyu  and
      Ferrara, Emilio",
    editor = "Demberg, Vera  and
      Inui, Kentaro  and
      Marquez, Llu{\'i}s",
    booktitle = "Proceedings of the 19th Conference of the {E}uropean Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = mar,
    year = "2026",
    address = "Rabat, Morocco",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-eacl/2026.eacl-long.348/",
    pages = "7416--7436",
    ISBN = "979-8-89176-380-7",
    abstract = "Personalization in LLMs often relies on costly human feedback or interaction logs, limiting scalability and neglecting deeper user attributes. We introduce (\textbf{G}enerative \textbf{R}esponse with \textbf{A}ligned \textbf{V}alues, \textbf{I}nterests, and \textbf{T}raits of \textbf{Y}ou), a framework for generating \textbf{synthetic, profile-grounded preference data} that captures users' interests, values, beliefs, and personality traits. By integrating demographic, cultural, and psychological frameworks{---}including Hofstede{'}s cultural dimensions, Schwartz{'}s basic values, the World Values Survey, and Big Five OCEAN traits{---}synthesizes chosen/rejected preference pairs to guide personalized content generation. We evaluate on book descriptions for 400 Amazon users, comparing it to prompt-based conditioning, standard fine-tuning, and naive synthetic pair generation. Profile-grounded synthetic data consistently improves generation, especially across multiple cultures (USA, Brazil, Japan, India), achieving over 4{\%} higher preference gains across baselines, with user studies showing that outputs are preferred over 86{\%} of the time. Our results show that scenario-grounded synthetic data can capture richer user variation, reduce reliance on costly annotation, and produce more engaging, user-centered content, offering a scalable path for LLM personalization. \textit{Code and datasets will be released upon acceptance.}"
}