Shaping Shared Languages: Human and Large Language Models' Inductive Biases in Emergent Communication

International Joint Conference on Artificial Intelligence, pp. 10298–10306

Abstract

Languages are shaped by the inductive biases of their users. Using a classical referential game, we investigate how artificial languages evolve when optimised for inductive biases in humans and large language models (LLMs) via Human-Human, LLM-LLM and Human-LLM experiments. We show that referentially grounded vocabularies emerge that enable reliable communication in all conditions, even when humans and LLMs collaborate. Comparisons between conditions reveal that languages optimised for LLMs subtly differ from those optimised for humans. Interestingly, interactions between humans and LLMs alleviate these differences and result in vocabularies more human-like than LLM-like. These findings advance our understanding of the role inductive biases in LLMs play in the dynamic nature of human language and contribute to maintaining alignment in human and machine communication. In particular, our work underscores the need to think of new LLM training methods that include human interaction and shows that using communicative success as a reward signal can be a fruitful, novel direction.

Authors 4

1. Tom Kouwenhoven Aachen Leiden Institute of Advanced Computer Science

   Leiden University

   Affiliation as printed

   Leiden Institute of Advanced Computer Science, Leiden University, Netherlands

2. Max Peeperkorn

   University of Kent

   Affiliation as printed

   School of Computing, University of Kent, United Kingdom

3. Roy de Kleijn Aachen Cognitive Psychology Unit Institute of Psychology, Leiden

   Leiden University

   Affiliation as printed

   Cognitive Psychology Unit, Institute of Psychology, Leiden, the Netherlands

4. Tessa Verhoef Aachen Leiden Institute of Advanced Computer Science

   Leiden University

   Affiliation as printed

   Leiden Institute of Advanced Computer Science, Leiden University, Netherlands