🌳 Tree of Thoughts: Problem Solving with Large Language Models
Failed to add items
Add to basket failed.
Add to wishlist failed.
Remove from wishlist failed.
Adding to library failed
Follow podcast failed
Unfollow podcast failed
🌳 Tree of Thoughts: Problem Solving with Large Language Models
-
Narrated by:
-
By:
About this listen
This research introduces Tree of Thoughts (ToT), a novel framework designed to enhance the problem-solving capabilities of large language models (LLMs). ToT moves beyond the token-by-token processing of existing methods like Chain of Thought by enabling LLMs to explore multiple coherent textual units, or "thoughts," as intermediate steps. This approach allows for deliberate decision-making through the consideration of various reasoning paths, self-evaluation of choices, and the ability to look ahead or backtrack. The authors demonstrate ToT's effectiveness on tasks requiring complex planning and search, such as the Game of 24, creative writing, and mini crosswords, achieving significant improvements over standard prompting techniques. The framework's modularity allows for flexibility in thought generation, evaluation, and search algorithms, offering a promising direction for more sophisticated LLM applications without the need for additional training.