ChatGPT Prompting Cannot Estimate Predictive Uncertainty in High-Resource Languages

TL;DR

This study evaluates ChatGPT's performance in high-resource languages and finds it struggles to accurately estimate its answer confidence, often being overconfident across multiple languages and NLP tasks.

Contribution

It provides the first analysis of ChatGPT's confidence calibration in high-resource languages, revealing overconfidence issues and performance similarities among these languages.

Findings

ChatGPT performs similarly across the five high-resource languages.

It exhibits poor confidence calibration, often overestimating its answer accuracy.

ChatGPT never assigns low confidence levels to its responses.

Abstract

ChatGPT took the world by storm for its impressive abilities. Due to its release without documentation, scientists immediately attempted to identify its limits, mainly through its performance in natural language processing (NLP) tasks. This paper aims to join the growing literature regarding ChatGPT's abilities by focusing on its performance in high-resource languages and on its capacity to predict its answers' accuracy by giving a confidence level. The analysis of high-resource languages is of interest as studies have shown that low-resource languages perform worse than English in NLP tasks, but no study so far has analysed whether high-resource languages perform as well as English. The analysis of ChatGPT's confidence calibration has not been carried out before either and is critical to learn about ChatGPT's trustworthiness. In order to study these two aspects, five high-resource languages and two NLP tasks were chosen. ChatGPT was asked to perform both tasks in the five languages and to give a numerical confidence value for each answer. The results show that all the selected high-resource languages perform similarly and that ChatGPT does not have a good confidence calibration, often being overconfident and never giving low confidence values.