A new study analyzed 300 personal narratives rewritten by three large language models, revealing a consistent pattern of stylistic normalization. The LLMs tended to decrease the use of function words, contractions, and first-person pronouns, while increasing vocabulary diversity and word length. These stylistic shifts occurred regardless of whether the prompt asked for general improvement or a simple rewrite, suggesting LLMs exert a pull towards a more polished, less situated writing style. AI
Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →
IMPACT LLM rewriting standardizes personal narratives, potentially impacting digital humanities and text analysis.
RANK_REASON Academic paper analyzing LLM capabilities and impact.