We gratefully acknowledge support from the Simons Foundation,

member institutions, and all contributors. Donate

Search... All fields Search

> cs > arXiv:2305.10601 Help | Advanced Search

Computer Science > Computation and Language

Access Paper:
[Submitted on 17 May 2023]
Download PDF
Tree of Thoughts: Deliberate Problem Solving with Large Language Models PostScript
Other Formats
Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L. Griffiths, Yuan Cao, Karthik Narasimhan
Current browse context:
Language models are increasingly being deployed for general problem solving across a wide range of tasks, but are still confined to token-level, cs.CL
left-to-right decision-making processes during inference. This means they can fall short in tasks that require exploration, strategic lookahead, < prev | next >
or where initial decisions play a pivotal role. To surmount these challenges, we introduce a new framework for language model inference, Tree of new | recent | 2305
Thoughts (ToT), which generalizes over the popular Chain of Thought approach to prompting language models, and enables exploration over Change to browse by:
coherent units of text (thoughts) that serve as intermediate steps toward problem solving. ToT allows LMs to perform deliberate decision making cs
cs.AI
by considering multiple different reasoning paths and self-evaluating choices to decide the next course of action, as well as looking ahead or
cs.LG
backtracking when necessary to make global choices. Our experiments show that ToT significantly enhances language models' problem-solving
abilities on three novel tasks requiring non-trivial planning or search: Game of 24, Creative Writing, and Mini Crosswords. For instance, in Game References & Citations
of 24, while GPT-4 with chain-of-thought prompting only solved 4% of tasks, our method achieved a success rate of 74%. Code repo with all NASA ADS
prompts: this https URL. Google Scholar
Semantic Scholar

Comments: Code repo with all prompts: this https URL 1 blog link (what is this?)
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) Export BibTeX Citation
Cite as: arXiv:2305.10601 [cs.CL]
Bookmark
(or arXiv:2305.10601v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2305.10601

Submission history
From: Shunyu Yao [view email]
[v1] Wed, 17 May 2023 23:16:17 UTC (609 KB)

Bibliographic Tools Code, Data, Media Demos Related Papers About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data
privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

About Contact Copyright Web Accessibility Assistance

Help Subscribe Privacy Policy arXiv Operational Status
Get status notifications via email or slack