# Predict the Next Word: Humans exhibit uncertainty in this task and language models _____

arXiv: https://arxiv.org/abs/2402.17527 · HTML: https://mlsanity.com/paper/2402.17527/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 58 references that are in the map.
- First posted on arXiv 27 Feb 2024.

## Builds on (most cited first, 30 of 34)

- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [BLOOM: A 176B-Parameter Open-Access Multilingual Language Model](https://mlsanity.com/paper/2211.05100/) (2022)
- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [When Does Label Smoothing Help?](https://mlsanity.com/paper/1906.02629/) (2019)
- [Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback](https://mlsanity.com/paper/2305.14975/) (2023)
- [Whose Opinions Do Language Models Reflect?](https://mlsanity.com/paper/2303.17548/) (2023)
- [Measuring Calibration in Deep Learning](https://mlsanity.com/paper/1904.01685/) (2019)
- [Beyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibration](https://mlsanity.com/paper/1910.12656/) (2019)
- [Reward learning from human preferences and demonstrations in Atari](https://mlsanity.com/paper/1811.06521/) (2018)
- [On Hallucination and Predictive Uncertainty in Conditional Language Generation](https://mlsanity.com/paper/2103.15025/) (2021)
- [Evaluating model calibration in classification](https://mlsanity.com/paper/1902.06977/) (2019)
- [What Can We Learn from Collective Human Opinions on Natural Language Inference Data?](https://mlsanity.com/paper/2010.03532/) (2020)
- [On Releasing Annotator-Level Labels and Information in Datasets](https://mlsanity.com/paper/2110.05699/) (2021)
- [Uncertainty Quantification with Pre-trained Language Models: A Large-Scale Empirical Analysis](https://mlsanity.com/paper/2210.04714/) (2022)
- [Calibration of Neural Networks using Splines](https://mlsanity.com/paper/2006.12800/) (2020)
- [Calibration tests in multi-class classification: A unifying framework](https://mlsanity.com/paper/1910.11385/) (2019)
- [Language Model Evaluation Beyond Perplexity](https://mlsanity.com/paper/2106.00085/) (2021)
- [Stop Measuring Calibration When Humans Disagree](https://mlsanity.com/paper/2210.16133/) (2022)
- [CLAM: Selective Clarification for Ambiguous Questions with Generative Language Models](https://mlsanity.com/paper/2212.07769/) (2022)
- [Calibration of Encoder Decoder Models for Neural Machine Translation](https://mlsanity.com/paper/1903.00802/) (2019)
- [On the Inference Calibration of Neural Machine Translation](https://mlsanity.com/paper/2005.00963/) (2020)
- [What Comes Next? Evaluating Uncertainty in Neural Text Generators Against Human Production Variability](https://mlsanity.com/paper/2305.11707/) (2023)
- [Can Large Language Models Capture Dissenting Human Voices?](https://mlsanity.com/paper/2305.13788/) (2023)
- [You should evaluate your language model on marginal likelihood over tokenisations](https://mlsanity.com/paper/2109.02550/) (2021)
- [Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models](https://mlsanity.com/paper/2305.19249/) (2023)
- [Tailoring Language Generation Models under Total Variation Distance](https://mlsanity.com/paper/2302.13344/) (2023)
- [Evaluating Computational Language Models with Scaling Properties of Natural Language](https://mlsanity.com/paper/1906.09379/) (2019)
- [Evaluating Distributional Distortion in Neural Language Modeling](https://mlsanity.com/paper/2203.12788/) (2022)

## Built on by (most cited first, 4 of 4)

- [A Survey on the Honesty of Large Language Models](https://mlsanity.com/paper/2409.18786/) (2024)
- [Asking a Language Model for Diverse Responses](https://mlsanity.com/paper/2509.17570/) (2025)
- [Variability Need Not Imply Error: The Case of Adequate but Semantically Distinct Responses](https://mlsanity.com/paper/2412.15683/) (2024)
- [Learning to vary: Teaching LMs to reproduce human linguistic variability in next-word prediction](https://mlsanity.com/paper/2509.17794/) (2025)

## Abstract (arXiv)

Language models (LMs) are statistical models trained to assign probability to human-generated text. As such, it is reasonable to question whether they approximate linguistic variability exhibited by humans well. This form of statistical assessment is difficult to perform at the passage level, for it requires acceptability judgements (i.e., human evaluation) or a robust automated proxy (which is non-trivial). At the word level, however, given some context, samples from an LM can be assessed via exact matching against a prerecorded dataset of alternative single-word continuations of the available context. We exploit this fact and evaluate the LM's ability to reproduce variability that humans (in particular, a population of English speakers) exhibit in the 'next word prediction' task. This can be seen as assessing a form of calibration, which, in the context of text classification, Baan et al. (2022) termed calibration to human uncertainty. We assess GPT2, BLOOM and ChatGPT and find that they exhibit fairly low calibration to human uncertainty. We also verify the failure of expected calibration error (ECE) to reflect this, and as such, advise the community against relying on it in this setting.