Название: Multilingual Artificial Intelligence
Автор: Peng Wang, Pete Smith
Издательство: Routledge
Год: 2025
Страниц: 164
Язык: английский
Формат: epub (true)
Размер: 10.1 MB
Multilingual Artificial Intelligence is a guide for non-computer science specialists and learners looking to explore the implementation of AI technologies to solve real-life problems involving language data.
Focusing on multilingual, multicultural, pre-trained large language models (LLMs) and their practical use through fine-tuning and prompt engineering, Wang and Smith demonstrate how to apply this new technology in areas such as information retrieval, semantic webs, and retrieval augmented generation, to improve both human productivity and machine intelligence. Finally, they discuss the human impact of language technologies in the cultural context, and provide an Artificial Intelligence (AI) competence framework for users to design their own learning journey.
With numerous generative pre-trained (GPT) models now available to the public, learners and general users can directly interact and experiment with AI tools and systems. These tools have become more accessible, helping people study, work, communicate, collaborate, and solve problems at a previously unseen scale involving language data. It is necessary to bring awareness of multilingual AI to a diverse audience, to readers and learners who will be able to use this technology in a responsible and ethical manner. With this book, our hope is to democratize and disseminate knowledge, as well as to develop skills and techniques in this ever-expanding and rapidly changing field.
From a computational perspective, a definitive element of human-level Artificial Intelligence in Deep Learning lies in its generalization capabilities, that is, its ability to produce sensible answers in response to new inputs that it never encountered during training. Deep Learning is “an approach to Machine Learning that involves training neural networks with many feed-forward layers on large datasets”. Deep Learning models in natural language applications are neural language models, which are used to convert a word symbol into a word vector (or word embedding) composed of learned semantic features in order to predict the next word in a sequence. Once trained, intermediate processing layers between the input and output layers can be thought of as representations of the training data with multiple levels of abstraction.
This innovative text is essential reading for all students, professionals, and researchers in language, linguistics, and related areas looking to understand how to integrate multilingual and multicultural Artificial Intelligence technology into their research and practice.
Скачать Multilingual Artificial Intelligence
[related-news]
[/related-news]