Skip to content
LLM Primer

zhou2022-least-to-most

arXiv: 2205.10625

TLDR (English)

"Break hard problems into easy ones, solve sequentially" is another reasoning paradigm parallel to CoT, especially effective for compositional generalization. Together with CoT/ToT forms the trio of "how to guide LLM step-by-step thinking".

TLDR（中文）

"先把难题拆成易题，再依次解决"是另一条与 CoT 平行的推理范式，对组合泛化任务尤其有效。和 CoT/ToT 一起构成"如何引导大模型分步思考"的三件套。