Philosophy and the Mind Sciences· 2026Q1
The Vector Grounding Problem
- 45citations
- Q1SCImago
- 2026year
Short summary
Large language models (LLMs) can achieve referential grounding, connecting their internal states to the real world, even without multimodal input or embodiment.
AI-generated from the title and abstract; the full text is not read.
Key points
- LLMs face a symbol grounding problem due to training solely on text, lacking direct world interaction.
- Referential grounding, the link between a representation and its real-world referent, is essential for LLMs to be about reality.
- Two conditions for referential grounding are proposed: appropriate causal-informational relations to the world and a history of selection for this function.
- LLMs can meet these conditions without needing multimodality or embodiment.
AI-generated from the title and abstract; the full text is not read.
Abstract
Large language models (LLMs) produce seemingly meaningful outputs, yet they are trained on text alone without direct interaction with the world. This leads to a modern variant of the classical symbol grounding problem in AI: can LLMs' internal states and outputs be about extra-linguistic reality, independently of the meaning human interpreters project onto them? We argue that they can. We first distinguish referential grounding—the connection between a representation and its worldly referent—from other forms of grounding and argue it is the only kind essential to solving the problem. We contend that referential grounding is achieved when a system's internal states satisfy two conditions derived from teleosemantic theories of representation: (1) they stand in appropriate causal-informational relations to the world, and (2) they have a history of selection that has endowed them with the function of carrying this information. We argue that LLMs can meet both conditions, even without multimodality or embodiment.
The authors' abstract, as published at the source. Philosophy and the Mind Sciences, 2026 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Experimental and Cognitive Psychology
Experimental and Cognitive PsychologyPsychology