LitVISTA: A Benchmark for Narrative Orchestration in Literary Text

2026-01-10

arXiv.org

AI Summary

This work addresses a systematic bias in current large language models, which overemphasize causal coherence at the expense of the intricate narrative arcs and structural orchestration characteristic of human storytelling when generating long-form narratives. To remedy this, the authors propose VISTA Space, a high-dimensional representation framework that unifies the modeling of narrative functions and structure, and introduce LitVISTA—the first structured evaluation benchmark tailored for literary narration. LitVISTA integrates high-dimensional narrative representations, fine-grained literary annotations, and an oracle-based assessment protocol leveraging mainstream large language models (e.g., GPT, Claude, Grok, Gemini) to systematically evaluate narrative structuring capabilities. Experiments reveal that existing models struggle to maintain globally coherent narrative perspectives and exhibit significant deficiencies in simultaneously capturing functional and structural narrative elements, with advanced reasoning strategies offering only marginal improvements.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Language and Vision

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Computational narrative analysis aims to capture rhythm, tension, and emotional dynamics in literary texts. Existing large language models can generate long stories but overly focus on causal coherence, neglecting the complex story arcs and orchestration inherent in human narratives. This creates a structural misalignment between model- and human-generated narratives. We propose VISTA Space, a high-dimensional representational framework for narrative orchestration that unifies human and model narrative perspectives. We further introduce LitVISTA, a structurally annotated benchmark grounded in literary texts, enabling systematic evaluation of models'narrative orchestration capabilities. We conduct oracle evaluations on a diverse selection of frontier LLMs, including GPT, Claude, Grok, and Gemini. Results reveal systematic deficiencies: existing models fail to construct a unified global narrative view, struggling to jointly capture narrative function and structure. Furthermore, even advanced thinking modes yield only limited gains for such literary narrative understanding.

Problem

• narrative orchestration
• literary text
• story arcs
• structural alignment
• large language models

Innovation

• narrative orchestration
• VISTA Space
• LitVISTA benchmark
• literary text analysis
• large language models