Tree of Thoughts
A prompt engineering method that extends single-pass reasoning into a multi-path sampling and voting mechanism. Regarded as an advanced variant of Chain-of-Thought prompting, it enables language models to explore multiple reasoning branches and aggregate results for more reliable outputs.
Core Facts
Mainstream task planning methods include the ReAct paradigm, Plan-and-Execute mode, and Tree of Thoughts
Inference-time computation improves output quality by allocating more computational resources during the inference stage while keeping model parameters fixed.
Timeline (last 90 days)
ReAct, Plan-and-Execute, and Tree of Thoughts are three different agent planning and task decomposition patterns, whose core difference lies in when and at what granularity planning occurs.
Mainstream task planning methods include the ReAct paradigm, Plan-and-Execute mode, and Tree of Thoughts
All Facts (7)
Mainstream task planning methods include the ReAct paradigm, Plan-and-Execute mode, and Tree of Thoughts
65%VerifiedInference-time computation improves output quality by allocating more computational resources during the inference stage while keeping model parameters fixed.
65%UnverifiedTree of Thoughts和Graph of Thoughts等学术研究正在推动深度Agent方向的发展
85%UnverifiedToT（Tree of Thoughts）在CoT基础上加入了搜索、评估和回溯机制，是目前最先进的推理框架之一
75%UnverifiedReAct has evolved into multiple variants such as Plan-and-Execute, Reflexion, and Tree of Thoughts.
60%UnverifiedThe mainstream implementation approaches for the thinking layer include the ReAct framework, Chain-of-Thought prompt chaining, and Tree of Thoughts methods.
60%UnverifiedReAct, Plan-and-Execute, and Tree of Thoughts are three different agent planning and task decomposition patterns, whose core difference lies in when and at what granularity planning occurs.
50%