2023
Tree of Thoughts: Deliberate Problem Solving with Large Language Models
More tokens and more latency per answer, so it earns its place only where a wrong answer is expensive. We do not reach for it on classification or extraction.
Interleaves verbal reasoning traces with tool actions, so the model thinks about what to do next, does it, then reasons about what came back. The reasoning trace also becomes an auditable record of why the agent took each step.
@inproceedings{yao2022react,
author = {Shunyu Yao and Jeffrey Zhao and Dian Yu and Nan Du and Izhak Shafran and Karthik Narasimhan and Yuan Cao},
title = {ReAct: Synergizing Reasoning and Acting in Language Models},
booktitle = {International Conference on Learning Representations (ICLR) 2023},
year = {2022},
eprint = {2210.03629},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2210.03629}
}Ejentic did not author this paper. Credit belongs to the authors named above; the note is ours.
2023
More tokens and more latency per answer, so it earns its place only where a wrong answer is expensive. We do not reach for it on classification or extraction.
2023
Cheap, and it works. When an agent fails a task, a written critique plus one retry often beats a bigger model.
2023
The benchmark we reach for when a client asks whether an agent can be trusted with a real repository. Passing tests is a much sharper signal than "the diff looks reasonable".