Adapting Large Language Models for Movie Domain with Narrative Understanding Tasks

Large language models (LLMs) have been deployed in a wide spectrum of domains and applications due to their strong language understanding capabilities obtained through pretraining.However, their performance on specific domain is usually suboptimal due to limited exposure to domain-specific tasks.Adapting LLMs to the movie domain poses unique challenges due to complex narratives that cannot be fully captured through subtitles or scripts alone.In this paper, we decompose movie understanding capability into a suite of narrative understanding tasks based on narrative theory.We construct a dataset for these tasks based on resources in the movie domain, and use it to examine the effect of different domain adaptation strategies.Our experiment results show the effectiveness of our approach in improving the narrative understanding of LLMs and highlight the trade-offs between domain-specific and general instruction capabilities.