# Large Language Models Preserve Semantic Isotopies in Story Continuations

arXiv: https://arxiv.org/abs/2510.04400 · HTML: https://mlsanity.com/paper/2510.04400/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 58 references that are in the map.
- First posted on arXiv 6 Oct 2025.

## Builds on (most cited first, 13 of 13)

- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [Large Language Models](https://mlsanity.com/paper/2307.05782/) (2023)
- [MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontiers](https://mlsanity.com/paper/2102.01454/) (2021)
- [A Corpus and Evaluation Framework for Deeper Understanding of Commonsense Stories](https://mlsanity.com/paper/1604.01696/) (2016)
- [Do LLMs write like humans? Variation in grammatical and rhetorical styles](https://mlsanity.com/paper/2410.16107/) (2024)
- [Adaptive Contrastive Search: Uncertainty-Guided Decoding for Open-Ended Text Generation](https://mlsanity.com/paper/2407.18698/) (2024)
- [A comparative evaluation and analysis of three generations of Distributional Semantic Models](https://mlsanity.com/paper/2105.09825/) (2021)
- [On the Efficacy of Sampling Adapters](https://mlsanity.com/paper/2307.03749/) (2023)
- [The Singleton Fallacy: Why Current Critiques of Language Models Miss the Point](https://mlsanity.com/paper/2102.04310/) (2021)
- [LongWanjuan: Towards Systematic Measurement for Long Text Quality](https://mlsanity.com/paper/2402.13583/) (2024)
- [Long-form analogies generated by chatGPT lack human-like psycholinguistic properties](https://mlsanity.com/paper/2306.04537/) (2023)
- [Instruction Tuning for Story Understanding and Generation with Weak Supervision](https://mlsanity.com/paper/2501.15574/) (2025)

## Abstract (arXiv)

In this work, we explore the relevance of textual semantics to Large Language Models (LLMs), extending previous insights into the connection between distributional semantics and structural semantics. We investigate whether LLM-generated texts preserve semantic isotopies. We design a story continuation experiment using 10,000 ROCStories prompts completed by five LLMs. We first validate GPT-4o's ability to extract isotopies from a linguistic benchmark, then apply it to the generated stories. We then analyze structural (coverage, density, spread) and semantic properties of isotopies to assess how they are affected by completion. Results show that LLM completion within a given token horizon preserves semantic isotopies across multiple properties.