Designing Large Language Model Applications

Transformer-based language models are powerful tools for solving a variety of language tasks and represent a phase shift in the field of natural language processing. But the transition from demos and prototypes to full-fledged applications has been slow. With this book, you'll learn the tools, techniques, and playbooks for building useful products that incorporate the power of language models.Experienced ML researcher Suhas Pai provides practical advice on dealing with commonly observed failure modes and counteracting the current limitations of state-of-the-art models. You'll take a comprehensive deep dive into the Transformer architecture and its variants. And you'll get up-to-date with the taxonomy of language models, which can offer insight into which models are better at which tasks.You'll learn:Clever ways to deal with failure modes of current state-of-the-art language models, and methods to exploit their strengths for building useful productsHow to develop an intuition about the Transformer architecture and the impact of each architectural decisionWays to adapt pretrained language models to your own domain and use casesHow to select a language model for your domain and task from among the choices available, and how to deal with the build-versus-buy conundrumEffective fine-tuning and parameter efficient fine-tuning, and few-shot and zero-shot learning techniquesHow to interface language models with external tools and integrate them into an existing software ecosystem

Suhas Pai is an experienced machine learning researcher, having worked in the tech industry for over a decade. He is the co-founder, CTO, and ML Research Lead at Bedrock AI, a Y-Combinator backed NLP startup operating in the financial domain. At Bedrock AI, Suhas invented several novel NLP techniques and LM based architectures that fully power the core features of Bedrock AIas products. Suhas is also the co-chair of the Privacy Working Group at Big Science for the BLOOM language model project, which when released was the world's largest open-source multilingual language model.Suhas is active in the ML community, being the Chair of the TMLS (Toronto Machine Learning Summit) ML conference in 2022, as well as the Chair of the TMLS NLP conference in 2021 and 2022. He is also the NLP lead at Aggregate Intellect, an ML community research organization, where he previously wrote a weekly AI newsletter summarizing and contextualizing the latest papers in ML (audience: 3k+) and hosted a weekly seminar to walk through recent NLP papers.