Embeddings and Retrieval for LLMs: Techniques and Challenges

Explore retrieval augmented generation techniques and challenges in this 35-minute conference talk by Anton Troynikov, cofounder of Chroma. Delve into the power of embedding-based retrieval for AI applications like 'chat with my documents' while gaining awareness of key limitations and mitigation strategies. Learn about Chroma, an open-source embedding database designed to simplify LLM app development by making knowledge, facts, and skills pluggable for large language models. Discover insights from Troynikov's background in robotics and 3D computer vision as he shares his perspective on the future of AI and its applications.

Embeddings and Retrieval for LLMs: Techniques and Challenges // Anton Troynikov // LLMs in Prod Con