Reproduce “Culture is Not Trivia: Sociocultural Theory for Cultural NLP” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Naitian Zhou, David Bamman, Isaac L. Bleaman
The paper argues for integrating sociocultural theory into cultural NLP, contending that current approaches often treat culture as static "trivia." It shows that "cultural facts" mined by NLP systems are frequently stereotypes and proposes a theoretical framework for understanding culture as dynamic and emergent, advocating for "localization" as a more pragmatic goal for system development.
Key insights
- Many current cultural NLP methods primarily learn "stereotypes" (higher-order indices) about cultural associations, rather than direct, first-order cultural practices.
- Culture is a dynamic, interactionally constructed, and context-dependent process, not a static collection of facts or "trivia."
- Sociocultural theory offers a conceptual toolkit to understand how identities and meanings are linked through linguistic forms and social contexts.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.