# Characterizing Narrative Content in Web-scale LLM Pretraining Data

arXiv: https://arxiv.org/abs/2606.19468 · HTML: https://mlsanity.com/paper/2606.19468/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 51 references that are in the map.
- First posted on arXiv 17 Jun 2026.

## Builds on (most cited first, 20 of 20)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Scaling Language Models: Methods, Analysis & Insights from Training Gopher](https://mlsanity.com/paper/2112.11446/) (2021)
- [Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research](https://mlsanity.com/paper/2402.00159/) (2024)
- [TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classification](https://mlsanity.com/paper/2010.12421/) (2020)
- [DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining](https://mlsanity.com/paper/2305.10429/) (2023)
- [RegMix: Data Mixture as Regression for Language Model Pre-training](https://mlsanity.com/paper/2407.01492/) (2024)
- [Art or Artifice? Large Language Models and the False Promise of Creativity](https://mlsanity.com/paper/2309.14556/) (2023)
- [Are Large Language Models Capable of Generating Human-Level Narratives?](https://mlsanity.com/paper/2407.13248/) (2024)
- [Automated Annotation with Generative AI Requires Validation](https://mlsanity.com/paper/2306.00176/) (2023)
- [Organize the Web: Constructing Domains Enhances Pre-Training Data Curation](https://mlsanity.com/paper/2502.10341/) (2025)
- [A Multi-Axis Annotation Scheme for Event Temporal Relations](https://mlsanity.com/paper/1804.07828/) (2018)
- [POTATO: The Portable Text Annotation Tool](https://mlsanity.com/paper/2212.08620/) (2022)
- [Narrative Planning: Balancing Plot and Character](https://mlsanity.com/paper/1401.3841/) (2014)
- [Trust No Bot: Discovering Personal Disclosures in Human-LLM Conversations in the Wild](https://mlsanity.com/paper/2407.11438/) (2024)
- [Knowledge Distillation in Automated Annotation: Supervised Text Classification with LLM-Generated Training Labels](https://mlsanity.com/paper/2406.17633/) (2024)
- [AboutMe: Using Self-Descriptions in Webpages to Document the Effects of English Pretraining Data Filters](https://mlsanity.com/paper/2401.06408/) (2024)
- [Where Do People Tell Stories Online? Story Detection Across Online Communities](https://mlsanity.com/paper/2311.09675/) (2023)
- [NarraBench: A Comprehensive Framework for Narrative Benchmarking](https://mlsanity.com/paper/2510.09869/) (2025)

## Built on by (most cited first, 2 of 2)

- [The Announcement Carries the Cue: Markup, Boundaries, and the Notation of Pre-Training Corpora](https://mlsanity.com/paper/2608.09093/) (2026)
- [Language Models Have Interpretable Narrative Mechanisms](https://mlsanity.com/paper/2610.12989/) (2026)

## Abstract (arXiv)

The narrative composition of web-scale LLM pretraining corpora remains largely unexplored, even though narrative is a fundamental mode of human communication. We present the first fine-grained study of narrative features in Dolma, a 3-trillion-token open pretraining corpus. Drawing on narrative theory, we design a framework spanning three core narrative elements (agency, setting, and events) operationalized as 11 interpretable dimensions. After curating and hand-annotating a diverse set of 400 passages, we create an LLM-labeled dataset of 25K passages, and finally, we finetune and validate NarraBERT, two RoBERTa-based models for fine-grained narrative prediction. We apply NarraBERT to 13M passages, resulting in a new dataset, NarraDolma. We find that narrative structure is measurable at scale across extremely heterogeneous data and narrative qualities are unequally distributed across pretraining sources, topics, and formats in ways that current data curation practices neither measure nor account for. Our framework, dataset, and analyses provide a foundation for understanding how narrative qualities are distributed in LLM pretraining data.