What Do Large Language Models Know? Tacit Knowledge as a Potential Causal-Explanatory Structure
This provides a conceptual framework for describing and explaining LLM behavior, addressing a foundational issue in AI and philosophy of mind, though it is incremental in applying existing philosophical concepts to LLMs.
The paper tackles the problem of understanding what knowledge large language models (LLMs) possess by arguing that LLMs can acquire tacit knowledge as defined by Martin Davies, despite his denial that neural networks can do so, and demonstrates that LLM architectural features meet constraints like semantic description and causal systematicity.
It is sometimes assumed that Large Language Models (LLMs) know language, or for example that they know that Paris is the capital of France. But what -- if anything -- do LLMs actually know? In this paper, I argue that LLMs can acquire tacit knowledge as defined by Martin Davies (1990). Whereas Davies himself denies that neural networks can acquire tacit knowledge, I demonstrate that certain architectural features of LLMs satisfy the constraints of semantic description, syntactic structure, and causal systematicity. Thus, tacit knowledge may serve as a conceptual framework for describing, explaining, and intervening on LLMs and their behavior.