"Understanding AI": Semantic Grounding in Large Language Models

TL;DR

This paper explores whether large language models understand meaning by examining their semantic grounding through philosophical theories, showing evidence of basic understanding across functional, social, and causal dimensions.

Contribution

It introduces a three-dimensional framework for semantic grounding and provides evidence that LLMs develop world models, challenging the view that they lack understanding.

Findings

LLMs show evidence of functional, social, and causal grounding.

LLMs develop world models indicating elementary understanding.

Grounding in LLMs is a gradual, multi-dimensional process.

Abstract

Do LLMs understand the meaning of the texts they generate? Do they possess a semantic grounding? And how could we understand whether and what they understand? I start the paper with the observation that we have recently witnessed a generative turn in AI, since generative models, including LLMs, are key for self-supervised learning. To assess the question of semantic grounding, I distinguish and discuss five methodological ways. The most promising way is to apply core assumptions of theories of meaning in philosophy of mind and language to LLMs. Grounding proves to be a gradual affair with a three-dimensional distinction between functional, social and causal grounding. LLMs show basic evidence in all three dimensions. A strong argument is that LLMs develop world models. Hence, LLMs are neither stochastic parrots nor semantic zombies, but already understand the language they generate, at least in an elementary sense.