[2502.18600] Chain of Draft: Thinking Faster by Writing Less

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status Get status notifications via email or slack

View PDF HTML (experimental) Abstract:Large Language Models (LLMs) have demonstrated remarkable performance in solving complex reasoning tasks through mechanisms like Chain-of-Thought (CoT) prompting, which emphasizes verbose, step-by-step reasoning. However, humans typically employ a more efficient strategy: drafting concise intermediate thoughts that capture only essential information. In this work, we propose Chain of Draft (CoD), a novel paradigm inspired by human cognitive processes, where LLMs generate minimalistic yet informative intermediate reasoning outputs while solving tasks. By…

saved by

• Towaki Takikawa

related reading

• [2402.12875] Chain of Thought Empowers Transformers to Solve Inherently Serial Problemsarxiv.org
• [2412.06769] Training Large Language Models to Reason in a Continuous Latent Spacearxiv.org
• [2201.11903] Chain of Thought Prompting Elicits Reasoning in Large Language Modelsarxiv.org
• Training Large Language Models to Reason in a Continuous Latent Spacearxiv.org
• Training Strategies for Efficient Embodied Reasoningarxiv.org
• [2305.04388] Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Promptingarxiv.org
• [2507.11473] Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safetyarxiv.org
• [2510.27338] Reasoning Models Sometimes Output Illegible Chains of Thoughtarxiv.org
• [2201.11903] Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelsarxiv.org
• [2205.11916] Large Language Models are Zero-Shot Reasonersarxiv.org
• LLM Post-Training: A Deep Dive into Reasoning Large Language Modelsarxiv.org
• [2510.24941] Can Aha Moments Be Fake? Towards Quantifying Decorative and True Thinking in Chain-of-Thoughtarxiv.org