# Narrative Lookahead in LLMs: Event-Level Decodability, Predictive Advantage, and Steering

arXiv: https://arxiv.org/abs/2610.18121 · HTML: https://mlsanity.com/paper/2610.18121/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 56 references that are in the map.
- First posted on arXiv 13 Oct 2026.

## Builds on (most cited first, 21 of 21)

- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Probing Classifiers: Promises, Shortcomings, and Advances](https://mlsanity.com/paper/2102.12452/) (2021)
- [Designing and Interpreting Probes with Control Tasks](https://mlsanity.com/paper/1909.03368/) (2019)
- [Plan-And-Write: Towards Better Automatic Storytelling](https://mlsanity.com/paper/1811.05701/) (2018)
- [Reasoning Models Know When They're Right: Probing Hidden States for Self-Verification](https://mlsanity.com/paper/2504.05419/) (2025)
- [Future Lens: Anticipating Subsequent Tokens from a Single Hidden State](https://mlsanity.com/paper/2311.04897/) (2023)
- [PlotMachines: Outline-Conditioned Generation with Dynamic Plot State Tracking](https://mlsanity.com/paper/2004.14967/) (2020)
- [Event Representations for Automated Story Generation with Deep Neural Nets](https://mlsanity.com/paper/1706.01331/) (2017)
- [Do language models plan ahead for future tokens?](https://mlsanity.com/paper/2404.00859/) (2024)
- [Activation Scaling for Steering and Interpreting Language Models](https://mlsanity.com/paper/2410.04962/) (2024)
- [Steering off Course: Reliability Challenges in Steering Language Models](https://mlsanity.com/paper/2504.04635/) (2025)
- [Unlocking the Future: Exploring Look-Ahead Planning Mechanistic Interpretability in Large Language Models](https://mlsanity.com/paper/2406.16033/) (2024)
- [Salience-Aware Event Chain Modeling for Narrative Understanding](https://mlsanity.com/paper/2109.10475/) (2021)
- [Memory and Knowledge Augmented Language Models for Inferring Salience in Long-Form Stories](https://mlsanity.com/paper/2109.03754/) (2021)
- [How Reliable are Causal Probing Interventions?](https://mlsanity.com/paper/2408.15510/) (2024)
- [STORYTELLER: An Enhanced Plot-Planning Framework for Coherent and Cohesive Story Generation](https://mlsanity.com/paper/2506.02347/) (2025)
- [Modeling Event Salience in Narratives via Barthes' Cardinal Functions](https://mlsanity.com/paper/2011.01785/) (2020)
- [Latent Planning Emerges with Scale](https://mlsanity.com/paper/2604.12493/) (2026)
- [Internal Planning in Language Models: Characterizing Horizon and Branch Awareness](https://mlsanity.com/paper/2509.25260/) (2025)
- [RAPTOR: Ridge-Adaptive Logistic Probes](https://mlsanity.com/paper/2602.00158/) (2026)
- [Where's the Plan? Locating Latent Planning in Language Models with Lightweight Mechanistic Interventions](https://mlsanity.com/paper/2605.07984/) (2026)

## Abstract (arXiv)

Do language models carry information about narrative events before writing them, and how far ahead? We examine narrative lookahead through salient, state-changing events rather than tokens or response-level attributes. We extract 23,355 LLM-annotated events from 2,000 stories generated by four open-weight LLMs, labeling them along three nonexclusive dimensions of change: within agents, between agents, and in the story world, with human validation on a subset. Event segmentation and plot-independent labels enable comparisons across stories at horizons measured in events rather than in tokens. At event ends, we probe hidden states for upcoming event types and compare predictions with text-only predictors using matched prefixes. Across four models, future event types are decodable above prevalence baselines for at least four subsequent events, but probes significantly outperform both primary text-only predictors only for the next event. In exploratory steering with OLMo-3.1-32B, a probe-derived direction shifts the proportion of the target event type in both intended directions at higher strengths. Our event-level approach links the prediction of future narrative developments to interventions that reshape subsequent event composition.