In recent years, large language models represented by ChatGPT and GPT-4 have made rapid technological progress and iteration, becoming the most revolutionary technology in the field of Artificial Intelligence. Large language models have made key breakthroughs in data information capacity, model parameter quantity, underlying model structure, and model training methods compared to previous language models. Their performance in tasks such as natural language processing, machine vision, and even general tasks continues to improve, including the emergence ability demonstrated by large language models. An overview of the technological evolution, architecture, key technologies, and main characteristics of large language models are provided in the paper. The basic architecture and core principles of large-scale models are introduced, their applications in the field of architecture are shared, their limitations, and future development directions are discussed. The aim is to promote the application and development of Artificial Intelligence technology represented by large language models in the field of architecture and civil engineering.
关键词
大型语言模型/涌现能力/适配调优/对齐/建筑行业大模型
Key words
large language models/emergent abilities/adaptation tuning/alignment/large model of architecture
维普
万方数据