Evaluating Pragmatic Reasoning in Large Language Models: Evidence from Scalar Diversity

TL;DR

This study investigates how different evaluation methods affect the assessment of pragmatic reasoning in large language models, revealing that results depend heavily on task design and model specifics.

Contribution

It compares probability measurement and metalinguistic prompting, showing that pragmatic reasoning varies across models and conditions, influenced by evaluation strategies.

Findings

Neither evaluation method consistently outperforms the other.

Pragmatic behavior varies across model families, prompting strategies, and task structures.

Scalar diversity gradients appear only in specific model-condition combinations.

Abstract

Evaluating pragmatic reasoning in large language models (LLMs) remains challenging because model behavior can vary depending on evaluation methods. Previous studies suggest that prompt-based judgments may diverge from models' internal probability distributions, raising questions about whether observed performance reflects underlying competence or task-induced behavior. This study examines this issue using scalar diversity as a graded diagnostic for pragmatic inference. Following Hu & Levy (2023), this study compares direct probability measurement and metalinguistic prompting across multiple models and experimental settings. The results show that neither evaluation method consistently outperforms the other and that pragmatic behavior varies substantially across model families, prompting strategies, and task structures. Moreover, scalar diversity gradients emerge only in specific model-condition combinations, suggesting that pragmatic reasoning in LLMs reflects an interaction between internal probabilistic representations and task-induced prompting behavior rather than a stable competence captured by a single evaluation paradigm. These findings highlight the central role of evaluation design in interpreting pragmatic abilities in LLMs.