Self-Improvement (for Language Models)

Presentation Link

Sam Witteveen and I were invited to speak at the main Singapore Data-Science meetup ("DSSG") for their July 2023 special on langchain.

My talk covered several topics under the broad heading of "Self-Improvement" (for Language Models):

• Improving Answers:
  • Self-Debug / Self-Refine
  • Constitutional AI
• Improving the System:
  • VOYAGER (GPT-4 planning for Minecraft)
• Creating training data:
  • phi-1
  • Evol-Instruct

The slides for my talk, which contain links to all of the reference materials and sources, are here :

If there are any questions about the presentation please ask below, or contact me using the details given on the slides themselves.