LLM narrative explanations don't boost human decision accuracy

A new study published on arXiv explores how persuasive narrative explanations from large language models (LLMs) affect human decision-making in classification tasks. Researchers found that while these explanations increased reliance on AI, they did not significantly improve decision accuracy compared to AI predictions alone. Furthermore, more persuasive narratives may have negatively impacted response times and the ability to discern correct AI predictions, suggesting potential trade-offs in using narrative explanations. AI

Summary written by gemini-2.5-flash-lite from 1 sources. How we write summaries →

IMPACT Investigates potential downsides of LLM explanations, suggesting careful consideration of their use in decision-making contexts.

RANK_REASON Academic paper on LLM capabilities. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• Large language models

• paper
• other