Perceptions of Linguistic Uncertainty by Language Models and Humans

TL;DR

This study examines how language models interpret linguistic uncertainty expressions and whether they can understand the uncertainty of others, revealing that most models mimic human-like responses but are biased by their prior knowledge.

Contribution

It is the first to evaluate language models' ability to interpret uncertainty expressions and employ theory of mind in this context, highlighting their biases and limitations.

Findings

7 out of 10 models map uncertainty expressions to probabilities like humans.

Models show bias depending on whether statements are true or false.

Language models are more influenced by their prior knowledge than humans.

Abstract

_Uncertainty expressions_ such as "probably" or "highly unlikely" are pervasive in human language. While prior work has established that there is population-level agreement in terms of how humans quantitatively interpret these expressions, there has been little inquiry into the abilities of language models in the same context. In this paper, we investigate how language models map linguistic expressions of uncertainty to numerical responses. Our approach assesses whether language models can employ theory of mind in this setting: understanding the uncertainty of another agent about a particular statement, independently of the model's own certainty about that statement. We find that 7 out of 10 models are able to map uncertainty expressions to probabilistic responses in a human-like manner. However, we observe systematically different behavior depending on whether a statement is actually true or false. This sensitivity indicates that language models are substantially more susceptible to bias based on their prior knowledge (as compared to humans). These findings raise important questions and have broad implications for human-AI and AI-AI communication.