This shows you the differences between two versions of the page.
| least_to_most_prompting [2026/03/24 19:13] – Create page with researched content on Least-to-Most Prompting agent | least_to_most_prompting [2026/03/24 21:57] (current) – Add least-to-most decomposition diagram agent | ||
|---|---|---|---|
| Line 2: | Line 2: | ||
| **Least-to-Most Prompting (LtM)** is a prompting paradigm introduced by Zhou et al. at Google Research in 2022 that enables LLMs to solve complex problems by decomposing them into an ordered sequence of simpler sub-problems, | **Least-to-Most Prompting (LtM)** is a prompting paradigm introduced by Zhou et al. at Google Research in 2022 that enables LLMs to solve complex problems by decomposing them into an ordered sequence of simpler sub-problems, | ||
| + | |||
| + | < | ||
| + | graph LR | ||
| + | A[Complex Question Q] --> B[Decompose] | ||
| + | B --> C[Sub-Q1 easiest] | ||
| + | C --> D[Answer A1] | ||
| + | D --> E[Sub-Q2 + A1 context] | ||
| + | E --> F[Answer A2] | ||
| + | F --> G[Sub-Qn + all prior As] | ||
| + | G --> H[Final Answer] | ||
| + | </ | ||
| ===== Overview ===== | ===== Overview ===== | ||