# States of LLM-generated Texts and Phase Transitions between them

arXiv: https://arxiv.org/abs/2503.06330 · HTML: https://mlsanity.com/paper/2503.06330/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 37 references that are in the map.
- First posted on arXiv 8 Mar 2025.

## Builds on (most cited first, 11 of 11)

- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [Hierarchical Neural Story Generation](https://mlsanity.com/paper/1805.04833/) (2018)
- [Learning to Write with Cooperative Discriminators](https://mlsanity.com/paper/1805.06087/) (2018)
- [Importance of Search and Evaluation Strategies in Neural Dialogue Modeling](https://mlsanity.com/paper/1811.00907/) (2018)
- [Criticality in Formal Languages and Statistical Physics](https://mlsanity.com/paper/1606.06737/) (2016)
- [Evaluating Computational Language Models with Scaling Properties of Natural Language](https://mlsanity.com/paper/1906.09379/) (2019)
- [Do Neural Nets Learn Statistical Laws behind Natural Language?](https://mlsanity.com/paper/1707.04848/) (2017)
- [Mutual Information Scaling and Expressive Power of Sequence Models](https://mlsanity.com/paper/1905.04271/) (2019)
- [Phase transition in large language models and the criticality of natural languages](https://mlsanity.com/paper/2406.05335/) (2024)
- [Autocorrelations Decay in Texts and Applicability Limits of Language Models](https://mlsanity.com/paper/2305.06615/) (2023)
- [Natural Language Statistical Features of LSTM-generated Texts](https://mlsanity.com/paper/1804.04087/) (2018)

## Built on by (most cited first, 3 of 3)

- [Temperature-driven inversion and nonlinear dynamics in ChatGPT-like AIs](https://mlsanity.com/paper/2608.00939/) (2026)
- [Estimating Text Temperature with Language Models](https://mlsanity.com/paper/2601.02320/) (2026)
- [Writing as a Self-Organized Critical Process](https://mlsanity.com/paper/2610.05466/) (2026)

## Abstract (arXiv)

It is known for some time that autocorrelations of words in human-written texts decay according to a power law. Recent works have also shown that the autocorrelations decay in texts generated by LLMs is qualitatively different from the literary texts. Solid state physics tie the autocorrelations decay laws to the states of matter. In this work, we empirically demonstrate that, depending on the temperature parameter, LLMs can generate text that can be classified as solid, critical state or gas.