Tag
This paper introduces NCP-Bench, a benchmark derived from 100 movie synopses for evaluating long-horizon narrative consistency in LLM-based interactive storytelling agents. Experiments show that even strong models like GPT-5.2 struggle to maintain logical consistency, with a 42% survival rate after 20 turns and high fact conflict rates.
This paper introduces NARRA-Gym, a benchmark and executable evaluation environment for assessing Large Language Models' abilities in sustaining interactive narratives, managing memory, and adapting to users over multiple turns.