Explore Airbnb's journey in evolving LLM fine-tuning support using Ray as its next-generation infrastructure component. Delve into the company's adoption of efficient fine-tuning techniques with Llama Factory and resource-sharing strategies using Kueue. Discover how Airbnb's customer support team leverages Supervised Fine Tuning (SFT) and Direct Preference Optimization (DPO) to align LLMs with curated preference datasets. Gain valuable insights into building and implementing advanced ML infrastructure, bridging the gap between theoretical concepts and practical applications in a major tech company. Learn from Shaowei Su and Mia Zhao as they present Airbnb's pivotal shift in AI capabilities during this 26-minute conference talk from the Ray Summit 2024.
Overview
Syllabus
Airbnb's LLM Evolution: Fine-Tuning with Ray | Ray Summit 2024
Taught by
Anyscale
Related Courses
-
Generative AI Advance Fine-Tuning for LLMs
-
Fine-Tuning LLM Models - Generative AI Course
-
Fine-tuning Language Models for Business Tasks
-
From Research to Production: Fine-Tuning and Aligning LLMs
-
FASTER Code for Supervised Fine-Tuning and DPO Training with UNSLOTH
-
Combined Preference and Supervised Fine-Tuning with ORPO
