SOTAVerified

Long-Form Narrative Summarization

Summarizing long-form narratives, such as books, movies, and TV scripts, remains an open challenge in NLP. Unlike news or document summarization, narratives require capturing intricate plotlines, evolving character relationships, and thematic coherence over tens of thousands of tokens. The hybrid structure of narratives, which combines descriptive prose with multi-speaker dialogues, implicit inference, and dynamic topic shifts, adds further complexity, demanding an approach that preserves contextual integrity while condensing information effectively. Furthermore, the sheer length of narrative texts, typically ranging from 40K to 160K tokens, poses significant challenges for standard summarization models.

Papers

Showing 18 of 8 papers

TitleStatusHype
NexusSum: Hierarchical LLM Agents for Long-Form Narrative Summarization0
Agent-as-Judge for Factual Summarization of Long NarrativesCode1
End-to-End Long Document Summarization using Gradient Caching0
MovieSum: An Abstractive Summarization Dataset for Movie ScreenplaysCode1
Chain of Agents: Large Language Models Collaborating on Long-Context Tasks0
Select and Summarize: Scene Saliency for Movie Script SummarizationCode0
BOOKSUM: A Collection of Datasets for Long-form Narrative Summarization0
BookSum: A Collection of Datasets for Long-form Narrative SummarizationCode1
Show:102550

No leaderboard results yet.