Why we built a self-writing Wikipedia
summary
Primer builds Quicksilver, a system that drafts first Wikipedia pages for missing scientists by linking names across news, Wikipedia and Wikidata and using structured claims extracted from articles. It trains on a corpus of 30,000 profiles provided with help from the Allen Institute, learns to resolve name ambiguity, generates human-readable summaries and an event stream of major developments, and publishes example output for a hundred scientists that editors can copy, verify and expand.
AI-written summary · about
chapters (4)
related talks
40mMigrating Freebase to Wikidata
19mA talk
35mYou've got questions. We've got answers! #bbuzz
59mExploring Wikipedia with Apache Spark: A Live Coding Demo
35mUnderstanding the News around the World with Web Scraping and NLP at Scale
46mWings Stay On or Replacing On Site Search at Wikipedia
39mHow to Avoid structural biases in evaluating your Machine Learning/NLP projects
48mBridging AI and HCI: Incorporating Human Values into the Development of AI Technologies