# Surprisal Minimisation over Goal-directed Alternatives Predicts Production Choice in Dialogue

arXiv: https://arxiv.org/abs/2605.00506 · HTML: https://mlsanity.com/paper/2605.00506/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 73 references that are in the map.
- First posted on arXiv 1 May 2026.

## Builds on (most cited first, 22 of 22)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference](https://mlsanity.com/paper/1704.05426/) (2017)
- [Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times?](https://mlsanity.com/paper/2212.12131/) (2022)
- [Testing the Predictions of Surprisal Theory in 11 Languages](https://mlsanity.com/paper/2307.03667/) (2023)
- [Revisiting the Uniform Information Density Hypothesis](https://mlsanity.com/paper/2109.11635/) (2021)
- [What Comes Next? Evaluating Uncertainty in Neural Text Generators Against Human Production Variability](https://mlsanity.com/paper/2305.11707/) (2023)
- [Structural Persistence in Language Models: Priming as a Window into Abstract Language Representations](https://mlsanity.com/paper/2109.14989/) (2021)
- [Generalized Measures of Anticipation and Responsivity in Online Language Processing](https://mlsanity.com/paper/2409.10728/) (2024)
- [A surprisal--duration trade-off across and within the world's languages](https://mlsanity.com/paper/2109.15000/) (2021)
- [Information Value: Measuring Utterance Predictability as Distance from Plausible Alternatives](https://mlsanity.com/paper/2310.13676/) (2023)
- [Do Language Models Exhibit Human-like Structural Priming Effects?](https://mlsanity.com/paper/2406.04847/) (2024)
- [Learning to refer informatively by amortizing pragmatic reasoning](https://mlsanity.com/paper/2006.00418/) (2020)
- [Surprise! Uniform Information Density Isn't the Whole Story: Predicting Surprisal Contours in Long-form Discourse](https://mlsanity.com/paper/2410.16062/) (2024)
- [An Incremental Iterated Response Model of Pragmatics](https://mlsanity.com/paper/1810.00367/) (2018)
- [Expectations over Unspoken Alternatives Predict Pragmatic Inferences](https://mlsanity.com/paper/2304.04758/) (2023)
- [Towards a Similarity-adjusted Surprisal Theory](https://mlsanity.com/paper/2410.17676/) (2024)
- [The Harmonic Structure of Information Contours](https://mlsanity.com/paper/2506.03902/) (2025)
- [Towards Pragmatic Production Strategies for Natural Language Generation Tasks](https://mlsanity.com/paper/2210.12828/) (2022)
- [Construction Repetition Reduces Information Rate in Dialogue](https://mlsanity.com/paper/2210.08321/) (2022)
- [Attribution and Alignment: Effects of Local Context Repetition on Utterance Production and Comprehension in Dialogue](https://mlsanity.com/paper/2311.13061/) (2023)
- [Is Information Density Uniform when Utterances are Grounded on Perception and Discourse?](https://mlsanity.com/paper/2602.14653/) (2026)

## Built on by (most cited first, 1 of 1)

- [Patterns of Priming in Production: Lexical, Semantic and Structural Alignment in Language Model Generation](https://mlsanity.com/paper/2609.04484/) (2026)

## Abstract (arXiv)

We model utterance production as probabilistic cost-sensitive choice over contextual alternatives, using information-theoretic notions of cost. We distinguish between goal-directed alternatives that realise a fixed communicative intent and goal-agnostic alternatives defined only by contextual plausibility, allowing us to derive speaker- and listener-oriented interpretations of different cost measures. We present a procedure to generate both types of alternative sets using language models. Analysing production choices in open-ended dialogue under both deterministic and probabilistic cost minimisation, we find that surprisal minimisation relative to goal-directed alternatives provides the strongest predictive account under both analyses. By contrast, uniform information density and length-based costs exhibit weaker and less consistent predictive power across conditions. More broadly, our study suggests that alternative-conditioned optimisation with LM-generated alternatives provides a principled framework for studying speaker and listener pressures in naturalistic language production.