Chain-of-density
Chain-of-density (CoD) is a prompting technique for summarisation that asks the model to iteratively produce denser summaries — each rewrite keeps the length but adds more entities.
Chain-of-density (Adams et al., 2023) had the model generate an initial summary, then rewrite it 4–5 times — each rewrite kept the same length but added missing entities from the source. The result: dense, information-rich summaries that consistently outperformed single-shot summarisation on entity coverage. In 2026 CoD is the default for long-document summarisation, news digests, and any case where the answer needs to be both compact and information-dense. Modern reasoning models partially handle this natively, but explicit CoD prompting still beats single-shot on dense factual material.
When to use chain-of-density
- Long-document summarisation with entity coverage requirements.
- News digest generation.
- Executive summary writing.
Common mistakes
- Stopping at iteration 2 — the gain shows up between iterations 3 and 5.
- Letting the length grow — the point is constant length with rising density.
FAQ
What is chain-of-density?
Chain-of-density (CoD) is a prompting technique for summarisation that asks the model to iteratively produce denser summaries — each rewrite keeps the length but adds more entities.
When should I use chain-of-density?
Long-document summarisation with entity coverage requirements. News digest generation. Executive summary writing.
What are the most common mistakes with chain-of-density?
Stopping at iteration 2 — the gain shows up between iterations 3 and 5. Letting the length grow — the point is constant length with rising density.
Related terms
- Chain-of-thought prompting — Chain-of-thought (CoT) prompting tells a language model to write its reasoning steps before its final answer, increasing accuracy on multi-step problems.
- Self-consistency — Self-consistency runs the same prompt multiple times at non-zero temperature and picks the most common final answer, raising accuracy on reasoning tasks.
- System prompt — A system prompt is the high-priority instruction block that defines a model's role, constraints, and default behaviors for an entire conversation.
Sources
Last updated: 2026-06-01. Raw markdown: https://promtable.com/glossary/chain-of-density.md.