LLMs can model story expectations, correlating with reader engagement

Researchers have developed a new framework using large language models (LLMs) to approximate consumers' expectations about stories. This method generates multiple imagined story continuations from a pre-trained LLM and extracts features like emotion and narrative path. The framework was validated through a survey-based approach comparing LLM-derived expectations to human beliefs and a rational-expectations approach comparing them to actual story outcomes. Findings indicate that LLM-derived expectations correlate with human beliefs and actual story continuations, and are associated with reader engagement. AI

IMPACT This framework offers a scalable method for understanding consumer beliefs about narrative content, potentially impacting content creation and platform strategy.

RANK_REASON The cluster contains an academic paper detailing a new framework for modeling story expectations using LLMs. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• George V. Guibas

• paper
• other

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