Large Knowledge Model: Perspectives and Challenges

arXiv:2312.02706v2 Announce Type: replace-cross
Abstract: Humankind's understanding of the world is fundamentally linked to our perception and cognition, with \emph{human languages} serving as one of the major carriers of \emph{world knowledge}. In this vein, \emph{Large Language Models} (LLMs) like ChatGPT epitomize the pre-training of extensive, sequence-based world knowledge into neural networks, facilitating the processing and manipulation of this knowledge in a parametric space. This article explores large models through the lens of "knowledge". We initially investigate the role of symbolic …

abstract arxiv challenges chatgpt cognition cs.ai cs.cl human knowledge language language models languages large language large language models llms major networks neural networks perception perspectives pre-training processing replace training type understanding world