Herding LLMs Towards Structured NLP

With the rise of the latest generation of large language models (LLMs), prototyping language processing (NLP) applications have become easier and more accessible than ever before. As amazing as these models are, pushing an LLM-centered application to a production-ready state comes with a new slate of challenges. This talk discusses spacy-llm - a library integrating LLMs into spaCy, leveraging its modular and customizable framework for working with text. This allows for a cheaper, faster and more robust NLP workflow from prototype to production - driven by cutting-edge LLMs, without compromising on having structured, validated data. Raphael Mitsch Machine learning engineer @ Explosion I'm a machine learning engineer with a MSc in data science. I have a soft spot for natural language processing, data visualization and the broader societal implications of AI. I've worked on numerous projects across different companies and industries, attempting to solve business problems using ML (occasionally even successfully). Currently working on spaCy at Explosion, where I dabble in LLMs, entity linking and other obscure NLP stuff. I think open and equitable access to AI-related knowledge and resources is crucial. I enjoy making things work, and I don't think attention is all we need.