On the Unexpected Abilities of Large Language Models

TL;DR

This paper reviews recent research on large language models, highlighting their unexpected cognitive abilities, the processes behind their development, and the implications for understanding AI and human cognition.

Contribution

It provides a comprehensive analysis of how LLMs develop indirect cognitive abilities and discusses factors enabling these abilities beyond their primary training objectives.

Findings

LLMs exhibit a wide range of unexpected cognitive skills.

The development of abilities is linked to indirect training processes.

Predicting the full capabilities of LLMs remains challenging.

Abstract

Large Language Models (LLMs) are capable of displaying a wide range of abilities that are not directly connected with the task for which they are trained: predicting the next words of human-written texts. In this article, I review recent research investigating the cognitive abilities developed by LLMs and their relation to human cognition. I discuss the nature of the indirect process that leads to the acquisition of these cognitive abilities, their relation to other indirect processes, and the implications for the acquisition of integrated abilities. Moreover, I propose the factors that enable the development of abilities that are related only very indirectly to the proximal objective of the training task. Finally, I discuss whether the full set of capabilities that LLMs could possibly develop is predictable.