Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

Stanford University

Exploring a Corpus - From Paper to Multilingual Chatbots

Stanford University via YouTube

Overview

Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
This 58-minute Stanford University workshop recording explores how the Stanford Open Virtual Assistant Lab created Genie, a system that transforms any text corpus into an interactive, multilingual chatbot. Learn how Genie uses a hybrid Retrieval-Augmented Generation (RAG) approach to minimize hallucinations by filtering LLM-generated text claim-by-claim against the source corpus. Discover how the system handles complex document formats like historical newspaper scans with scattered article layouts. Examine impressive case studies including Wikipedia implementations in 25 languages achieving 98% factual accuracy (compared to GPT-4's 43%), The African Times newspaper from the 19th century, and the Chronicling America historical newspaper collection. Presented by Sina Semnani on February 14, 2025, this invitation-only workshop was sponsored by the Alfred P. Sloan Foundation and Stanford HAI, focusing on public AI assistants for worldwide knowledge and their implications for the free web.

Syllabus

Public AI Assistant to Worldwide Knowledge: Exploring a Corpus – From Paper to Multilingual Chatbots

Taught by

Stanford HAI

Reviews

Start your review of Exploring a Corpus - From Paper to Multilingual Chatbots

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.