Writing

Text Summarization Using the Chain of Density (CoD) Method

Given an article, the Chain of Density (CoD) method guides generating progressively concise summaries that densely include relevant entities. Starting with a verbose, minimally informative summary of about 80 words, the process involves iteratively identifying informative, concis

6 steps 1 variables English

Prompt template

Run these steps in order.

01
Input the full article text into ARTICLE.
02
Write an initial long (~80 words) nonspecific summary using verbose language and minimal informative entities.
03
Repeat the following five times:
04
Identify 1-3 relevant, concise (five words or fewer), novel, and faithful entities missing from the previous summary.
05
Rewrite the summary to include all previous and new entities, maintaining the same word count, improving flow, and maximizing information density.
06
Ensure each summary is self-contained and retains all prior entities while balancing new entities within the length constraint.

Prompt library

Use these prompts directly inside ChatGPT.

Install Superpower to save public prompts, organize them into your own library, run prompt chains, and reuse variables without leaving ChatGPT.