Enhancing LLMs with Structured Outputs and Function Calling

About

Dive deep into the capabilities of Large Language Models (LLMs) with a focus on structured outputs, which are crucial for advanced data manipulation and application-specific customization. This session will explore the tools and techniques for interacting with and optimizing LLMs, highlighting practical strategies for fine-tuning and leveraging structured data formats like dictionaries and JSON. Participants will gain hands-on insights into enhancing the functionality of both proprietary and open-sourced LLMs through structured outputs and function calling.

Key Takeaways:

  • Foundational Understanding of LLMs: Learn what LLMs are, their main applications, and how they can be customized through structured outputs.
  • Tools for Structured Interaction: Get acquainted with Python frameworks like Pydantic and Instructor for effective LLM interactions.
  • Enhancing LLMs for RAG: Understand the role of structured outputs in Retrieval-Augmented Generation applications to improve data retrieval and utilization.
  • Function Calling Techniques: Discover how to implement function calling in LLMs to automate and enhance processing tasks.
  • Optimization Strategies: Explore methodologies for the continuous evaluation and optimization of LLMs using tools like W&B Weave for better performance monitoring.

Speaker

Book Tickets
Stay informed about DHS 2025

Download agenda

We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy & Cookies Policy.

Show details