Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
Chain-of-thought prompting shows that step-by-step examples unlock multi-step reasoning in large models — no training changes, just better prompts.
5 steps · about 45 min · 2022–2023
Prompting as a reasoning interface: chain-of-thought, self-consistency, tool-using ReAct agents and deliberate search over thoughts.
Prefer to jump around? Each step is a standalone report with its own citations, figures and FAQ.
Chain-of-thought prompting shows that step-by-step examples unlock multi-step reasoning in large models — no training changes, just better prompts.
An AI-generated review of Self-Consistency decoding, showing how sampling multiple reasoning paths and selecting the most consistent answer improves.
ReAct interleaves chain-of-thought reasoning with tool actions like search and lookup, grounding language models in external feedback.
Tree of Thoughts explores multiple reasoning branches with search and self-evaluation, solving problems where single-pass chain-of-thought fails.
A synthetic proposal for causal attention: separating correlational co-occurrence from causal influence so transformers can reason over interventions.