ML+Coffee: Retrieval Augmented Generation (RAG) & SQLDeps

ML+Coffee offers a supportive and casual environment to discuss ongoing machine learning (ML) / AI projects and share knowledge & tools across campus. No matter your background, you'll find a welcoming space to connect, learn, and share insights with the ML+X community.

Registration: Please register to help us plan our catering needs.

Discussion(s) and Demo(s) in Queue

1. 9:00-9:30am: SQLDeps Demo (Cainã Max Couto-Silva) https://github.com/glue-lab/sqldeps 
2. 9:30-10:00am: RAG Demo (Chris Endemann): Learn how to build a simple RAG pipeline in Python.
3. 10:00-11:00am: RAG Discussions (Ron Stewart): We have been exploring whether we can evaluate PubMed abstracts using LLM analysis with Retrieval Augmented Generation based on co-occurrence models to provide some level of support or refutation for a hypothesis more quickly. It would be great to get feedback on not only the ML analysis, but the analysis in general, and also there are some issues with the statistics that we still haven't worked out.