Tree of Thoughts: Deliberate Problem Solving with Large Language ModelsTree of Thoughts: Deliberate Problem Solving with Large Language Models
问题:LM 推理仍是 token 级、从左到右的决策,在需要探索、策略前瞻或初始决策起关键作用的任务上会失效;能否让 LM 在以「思维」为中间步骤的树上做 deliberate 搜索——考虑多条推理路径、自评估选择、必要时前瞻与回溯。
证据位置:摘要给出 Game of 24 上 GPT-4 + CoT 仅解出 4% 而 ToT 达 74%;实验在 Section 4.1 Table 2——从 4nums.com 抓取的 1,362 局中取较难的第 901-1,000 局、以 100 局成功率为指标,GPT-4、温度 0.7(实验于 2023 年 5 月 5-16 日进行);成本与局限在 Section 6 Discussion 与 Appendix B.3。
作者自陈局限:Section 6 写明这类 deliberate search "might not be necessary for many existing tasks that GPT-4 already excels at",且作为初步工作只探索了三个相对简单的任务;Appendix B.3 写明 "could require 5-100 times more generated tokens than CoT";工作只用现成 LM,微调被留作未来方向。
它没主张的:74% 是 100 局较难子集、BFS 每步保留 b=5 候选设置下的成功率,不是全部 1,362 局的成绩;论文没有证明 ToT 处处优于 CoT——Appendix B.1 在 GSM8K/StrategyQA 上只有小幅提升,B.2 中 GPT-3.5+ToT 在 Game of 24 仅 19%;作者也指出一般填字游戏可由专用 NLP 管线解决,本文并未主张取代它们,更没有任何训练层面的结论。Problem: LM inference remains token-level, left-to-right decision making, which falls short on tasks requiring exploration, strategic lookahead, or where initial decisions play a pivotal role; can an LM instead search deliberately over a tree with thoughts as intermediate steps — considering multiple reasoning paths, self-evaluating choices, and looking ahead or backtracking when necessary.
Where the evidence sits: the abstract reports 74% success for ToT vs 4% for GPT-4 with chain-of-thought prompting on Game of 24; the experiment is in Section 4.1, Table 2 — the relatively hard games indexed 901-1,000 out of 1,362 scraped from 4nums.com, success rate over 100 games, GPT-4 at temperature 0.7 (experiments run May 5-16, 2023); cost and limits are in Section 6, Discussion, and Appendix B.3.
What the authors say about limits: Section 6 states that deliberate search "might not be necessary for many existing tasks that GPT-4 already excels at" and that, as an initial step, only three relatively simple tasks are explored; Appendix B.3 states ToT "could require 5-100 times more generated tokens than CoT"; the work uses an off-the-shelf LM, with fine-tuning left as a future direction.
What it does not claim: 74% is the success rate on the 100-game hard subset with BFS keeping b=5 candidates per step, not a result on all 1,362 games; the paper does not show ToT beats CoT everywhere — Appendix B.1 reports only slight gains on GSM8K/StrategyQA and Appendix B.2 reports GPT-3.5+ToT at 19% on Game of 24; the authors note general crosswords can be solved by specialized NLP pipelines, and no training-level claim is made.
关键参数Key specs
- 证据位置 / evidence
- 摘要;Section 4.1 Table 2(Game of 24);Appendix B.3 成本分析
- Game of 24 成功率 / success
- ToT 74% vs GPT-4 CoT 4%
- 测试集 / test set
- 4nums.com 共 1,362 局,取较难的第 901-1,000 局,以 100 局成功率为指标
- 额外生成开销 / cost
- 生成 token 可达 CoT 的 5-100 倍
- 复现状态 / reproduction
- 未复现
官方原文怎么说What the source says
For instance, in Game of 24, while GPT-4 with chain-of-thought prompting only solved 4% of tasks, our method achieved a success rate of 74%.
ToT allows LMs to perform deliberate decision making by considering multiple different reasoning paths and self-evaluating choices to decide the next course of action, as well as looking ahead or backtracking when necessary to make global choices.
We scrape data from 4nums.com, which has 1,362 games that are sorted from easy to hard by human solving time, and use a subset of relatively hard games indexed 901-1,000 for testing. For each task, we consider the output as success if it is a valid equation that equals 24 and uses the input numbers each exactly once. We report the success rate across 100 games as the metric.
In general, cost and efficiency of ToT highly depend on the prompts and search algorithms used, and could require 5-100 times more generated tokens than CoT.
资料与许可Source and licence
- 资料时间Source checked
- 2026-11-07
- 许可 / 类型Licence / kind
- arxiv-preprint
- 许可说明Licence evidence
- https://arxiv.org/abs/2305.10601
- 类目Category
- 推理Reasoning
论文里的数字怎么读How to read the numbers
先分清测什么、怎么算、和谁比、有没有数据污染。拿不准就回原文看实验设置。Before citing a score, separate what it measures, how it is scored, what it compares against and whether contamination is controlled. If unsure, go back to the experiment setup in the paper. 看评测基准See benchmark cards · 看数据集卡See dataset cards