Also known as: Chain-of-Thought · COT
COT, o chain of thought, se refiere a una secuencia de pasos intermedios de razonamiento utilizados o solicitados al resolver una tarea. En IA generativa, el término suele describir prompting orientado al razonamiento o comportamiento del modelo destinado a mejorar el rendimiento en problemas de varios pasos.
In generative AI, Chain-of-Thought (CoT) is a prompting technique that encourages the large language model (LLM) to explicitly detail its reasoning process before arriving at a conclusion. This involves prompting the model to show the intermediate steps it takes to solve a problem, rather than just providing the final answer. This method can significantly improve the LLM's performance on complex reasoning tasks.