Large Language Models (LLMs) are often criticized for lacking true
"understanding" and an ability to "reason" with their knowledge, being seen
merely as advanced autocomplete systems. We believe that this perspective might
be missing an important insight. We suggest that LLMs do develop a kind of
empirical "understanding" that is "geometry"-like, which seems quite sufficient
for a range of applications in NLP, computer vision, coding assistance, etc.
However, this "geometric" understanding, built from incomplete and noisy data,
makes them unreliable, difficult to generalize, and lacking in inference
capabilities and explanations, similar to the challenges faced by
heuristics-based expert systems decades ago.
To overcome these limitations, we suggest that LLMs should be integrated with
an "algebraic" representation of knowledge that includes symbolic AI elements
used in expert systems. This integration aims to create large knowledge models
(LKMs) that not only possess "deep" knowledge grounded in first principles, but
also have the ability to reason and explain, mimicking human expert
capabilities. To harness the full potential of generative AI safely and
effectively, a paradigm shift from LLMs to the more comprehensive LKMs is
needed.

大型语言模型 (LLMs) 通常被批评缺乏真正的 “理解” 和 “推理” 能力，被视为高级自动填充系统。本研究认为这种观点可能忽视了一个重要的见解，即 LLMs 确实发展出一种类似于 “几何” 的经验性 “理解”，这对自然语言处理、计算机视觉、代码辅助等领域的应用似乎足够了。然而，这种基于不完全和嘈杂数据构建的 “几何” 理解使得 LLMs 不可靠、难以推广，并且缺乏推理能力和解释能力，类似于几十年前启发式专家系统面临的挑战。为了克服这些局限性，本研究建议将 LLMs 与包括专家系统中使用的符号 AI 元素在内的 “代数” 知识表示集成起来。这种集成旨在创建大型知识模型 (LKMs)，它们不仅具备以第一原理为基础的 “深度” 知识，而且具备推理和解释的能力，模仿人类专家的能力。为了安全有效地利用生成式 AI 的全部潜力，需要从 LLMs 转向更全面的 LKMs，这是一种新的范式转变。