Recent searches
Search options
Administered by:
Chain of Draft: LLMs Thinking Faster with Less Text 
#Zoom researchers introduce #ChainofDraft (#CoD), a new #LLM prompting method that achieves comparable accuracy to #ChainofThought while using only 7.6% of the tokens, significantly reducing cost and latency.
#ChainofDraft is inspired by human problem-solving - using concise drafts instead of verbose reasoning
Matches or exceeds #ChainofThought accuracy across multiple reasoning tasks
Reduces token usage by up to 92.4% compared to verbose methods
Significantly lowers computational costs for complex reasoning
Works with leading #LLMs like GPT-4o and Claude 3.5 Sonnet
Excels in arithmetic, common-sense, and symbolic reasoning benchmarks
arXiv.orgChain of Draft: Thinking Faster by Writing LessLarge 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 reducing verbosity and focusing on critical insights, CoD matches or surpasses CoT in accuracy while using as little as only 7.6% of the tokens, significantly reducing cost and latency across various reasoning tasks. Our code and data are available at https://github.com/sileix/chain-of-draft.