Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
arXiv: 2201.11903
TLDR(中文)
提出 Chain-of-Thought(思维链)提示技术:通过在提示中加入中间推理步骤, 可以大幅提升大语言模型在数学、逻辑、常识推理等任务上的表现。 这个简单技巧把 LLM 的推理能力推向了接近人类的水平。
TLDR (English)
Introduces chain-of-thought prompting: adding intermediate reasoning steps to prompts dramatically improves LLM performance on math, logic, and commonsense reasoning tasks. This simple technique brought LLM reasoning capabilities close to human-level performance.