Estimating Text Temperature with Language Models
Not on the map. Why?
Abstract (arXiv)
Autoregressive language models typically use temperature parameter at inference to shape the probability distribution and control the randomness of the text generated. After the text was generated, this parameter can be estimated using maximum likelihood approach. Following it, we propose a procedure to estimate the temperature of any text, including ones written by humans, with respect to a given language model. We evaluate the temperature estimation capability of a wide selection of small-to-medium Large Language Models (LLMs). We then use the best-performing Qwen3 14B to estimate temperatures of popular corpora, finding that while most measured temperatures are close to 1, notable exceptions include Jokes, GSM8K, and AG News (1.1), and Python code (0.9).
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 7 of its 21 references that are in the map. First posted on arXiv 5 Jan 2026.
Builds on
7 of 21 references.
References · 7
- 2021Training Verifiers to Solve Math Word Problemscited by 10,109
- 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
- 2016Pointer Sentinel Mixture Modelscited by 3,096
- 2019The Curious Case of Neural Text Degenerationcited by 2,916
- 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
- 2025Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAscited by 521
- 2025States of LLM-generated Texts and Phase Transitions between themcited by 3
Built on by
Not cited yet.