LLM use may create linguistic monoculture, new paper warns

A new paper proposes a mathematical framework to understand how widespread use of large language models (LLMs) in writing could lead to a "linguistic monoculture." The research models authors and LLMs as distributions of linguistic features that coevolve through interaction. It analyzes scenarios with fixed shared models, recursively updated models, and personalized models, finding that shared models can reduce linguistic diversity, while personalization can maintain it. The study suggests that individuals may conform more than is socially optimal, creating a "price of monoculture" due to a negative externality where distinctiveness is undervalued. AI

IMPACT May lead to reduced linguistic diversity in written communication as LLMs become more integrated into writing processes.

RANK_REASON The cluster contains a single academic paper discussing a theoretical framework for LLM impact on language diversity. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• arXiv
• CatalyzeX
• cs.CL
• DagsHub
• Gotit.pub
• Hugging Face
• Influence Flower
• linguistic monoculture
• LLM
• ScienceCast

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →