prompting · chain-of-thought TEC-0003

Show intermediate steps

Prompt the model to write out intermediate reasoning steps before the final answer.

Few-shot examples or a simple instruction that elicits step-by-step reasoning improve multi-step arithmetic and symbolic tasks in large models. Modern reasoning-tuned models do this by default, so the gain is largest on older or smaller models.

Addresses: Digit-level arithmetic

Contexts: Math, General

Does it work?

contested2 supporting · 1 contesting sources · last moved 2026-09-04

Efficacy claims — what this technique actually moves, under which conditions, and whether that has been contested.

Fails when:

The error is inside a step's arithmetic rather than in the decomposition. More steps do not fix it; the computation has to leave the text. (helps, narrowly)

mechanismreplicatedcontested
Writing out reasoning steps improves how a problem is decomposed but does not fix the arithmetic inside a step — the model can decompose correctly and still get the calculation wrong, at the same confident register as when it is right. For exact arithmetic the fix is to move the computation out of the text, not to add more steps.
- supports PAL: Program-aided Language Models (2022)
- supports Faith and Fate: Limits of Transformers on Compositionality (2023)
- contests Tokenization counts: the impact of tokenization on arithmetic in frontier LLMs (2024)

Code

No repository linked yet. Contribute one.

Sources

Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (2022)

Either way, bring evidence: a link, or what you saw.

Whether a technique works is a claim. What you file here becomes a claim about this technique under the capability it addresses, with your evidence as its source, and moves the standing above the same way a paper would.