Last week, researchers from Google #Deepmind and Princeton University released a paper titled: “Tree of Thoughts: Deliberate Problem Solving with Large Language Models”, introducing a new way to overclock Large Language Models and make them solve problems like humans.
In this video, I cover the key takeaways from the paper, briefly explaining what Tree of Thought promoting is and other prompting techniques like Chain of Thought, Input-Output and Chain of Thought with Self Reflection as used in Google’s PaLM 2 Technical Report, and the experiments done by research to show just how much smarter Tree of Thought is.
Relevant Links and References:
Tree of Thought Paper:
https://arxiv.org/pdf/2305.10601.pdf
Google PaLM 2 Technical Report:
https://ai.google/static/documents/palm2techreport.pdf
@WesRoth Explainer:
https://youtu.be/BrjAt-wvEXI
Google Deepmind:
https://www.deepmind.com
Chatpters:
Intro - 0:00
What Is Tree of Thought - 0:27
What is Input-Output Prompting - 0:56
What is Chain of Thought Prompting - 1:10
What is Chain of Thought with Self Reflection - 1:37
How Does Tree of Thought Work - 2:16
Tree of Thought Experiments - 3:25
Conclusion - 5:54
Thanks for watching, have a great weekend.
#AI #Trending #AIResearch #LLM #Google
Share this page with your family and friends.