Today, during my presentation on LLM evaluation at the SF Big Analytics and AICamp event, we had a productive discussion. The presentation deck is available here. Attendees posed insightful questions regarding the evaluation of search, retrieval, and question-answering systems, particularly those employing hybrid architectures such as systems composed of Retrieval-Augmented Generation (RAG), LLMs integrated with search tools inspired by FreshPrompt/FreshLLMs, and graph-based LLM search. Additionally, there was significant interest in the evaluation of multimodal systems, which present unique complexities. This topic is especially critical as multimodal search and applications, such as applications in healthcare and medicine, gain traction.
Rohit Aggarwal
5mo
Thank you for sharing.
Malathi Sankar
4mo
This is super informative Andrei Thank you for sharing!
Product Manager | GenAI (LLMs, RAGs, Agents), AdTech, SaaS & eCommerce products
5moAndrei Lopatenko 🇺🇦 is there a recording? sounds awesome!