Modeling Cultural and Subcultural Variation in Code-Switched Discourse with Topic Annotation

2026-06-25 · Source: Paper Index on ACL Anthology · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Expert, quick

Summary

Nemika Tyagi, Nelvin Licona-Guevara, and Olga Kellert propose topic-based annotation as a framework to analyze cultural and subcultural variation in bilingual discourse, moving beyond structural or token-level code-switching models. Their work utilizes large language models to annotate 3,691 code-switched sentences from Spanish-English (Miami) and Spanish-Guaraní (Paraguay) corpora, integrating topic, discourse-level information, and sociolinguistic metadata. The analysis reveals systematic relationships among discourse topics, language choice, and social variables like gender and language dominance. Key observations include subcultural variation within the Miami community and a distinct diglossic distribution in Paraguay, where Guaraní is associated with formal domains and Spanish with informal communication. These findings indicate that modeling code-switching through discourse-level categories provides a more comprehensive representation of multilingual communication, facilitating both cross-cultural and intra-cultural comparisons at scale.

Key takeaway

For NLP Engineers and Research Scientists developing models for multilingual discourse, you should integrate discourse-level categories and sociolinguistic metadata into your analysis. This approach, demonstrated by topic-based annotation, offers a more complete representation of code-switching, revealing cultural and subcultural variations. Consider moving beyond token-level analysis to capture the social and cultural nuances that shape language choice, enabling more accurate and context-aware multilingual communication systems.

Key insights

Modeling code-switching via discourse topics and social context reveals cultural and subcultural variations in bilingual communication.

Principles

• Code-switching is a discourse practice shaped by social context.
• Discourse-level categories enable cross-cultural comparison.
• Language choice correlates with social variables.

Method

Annotate code-switched sentences with topic and discourse-level information using large language models, integrating sociolinguistic metadata for analysis.

In practice

• Analyze code-switching beyond token-level phenomena.
• Identify diglossic distributions in bilingual communities.
• Compare language use across cultural subgroups.

Topics

• Code-switching
• Topic Annotation
• Large Language Models
• Bilingual Discourse
• Sociolinguistics
• Cultural Variation

Best for: AI Scientist, NLP Engineer, Research Scientist

Related on AIssential

• Decoding the Maghreb Diaspora: A Multilingual NLP Framework for Social Integration — Specialized NLP architectures are crucial for handling code-switching in under-resourced, hybrid languages.
• A Cross-Genre Analysis of Discourse Relation Signaling in the GUM Corpus — The paper analyzes discourse relation signaling across diverse genres in the GUM Corpus.
• From 50K to 8.2 Million in 24 Hours: Vozinha's Algorithmic Consecration and the Multilingual Making of World Cup Visibility — Algorithmic visibility for public figures is shaped by distinct cross-lingual narrative frames and platform metrics.
• Annotating Nigerian English for AI: Practical Insights from a Native Lagos-Based Annotator — Nigerian English annotation requires native expertise to prevent AI model inaccuracies and cultural misrepresentation.
• Evaluating Cross-lingual Knowledge Consistency in Code-Mixed vis-a-vis Indian Languages using IndicKLAR — Code-mixing inputs significantly improves large language model knowledge recall in Indian languages, nearly matching English performance.
• A corpus-assisted metaphor analysis in portraying businesswomen: Diachronic changes in Hong Kong English news — Corpus-assisted metaphor analysis reveals diachronic shifts in media portrayals of businesswomen.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Paper Index on ACL Anthology.