Introduction to Large Language Models
摘要
This chapter introduces the reader to the domain of artificial intelligence, machine learning and deep learning to give a high-level understanding of how LLMs are designed, trained and fine-tuned. It explores the variety of tasks these models can solve: from generating text following user instructions to estimating sentences’ semantic similarity (how similar in terms of meaning two sentences are); from text summarisation to content translation and much more. The chapter closes with current known limitations and how the research community and the industry are working to mitigate them.