用轻量提示工程将专利转化为三年内可落地的产品创意。
MK2 at PBIG Competition: A Prompt Generation Solution
- 以提示为中心的迭代优化流程，融合弱输出中的有效片段。
- 在三大领域六项指标中胜出25/36测试，自动评测领先。
- 无需额外训练数据，适合快速生成商业级创新想法。
专利驱动的创意生成任务要求系统将真实专利转化为三年内可实现的产品创意。我们提出MK2，一种以提示为核心的流水线：Gemini 2.5负责起草并迭代修改提示，从较弱输出中提取有用片段；GPT-4.1则利用该提示为每篇专利生成一个创意；最后由Qwen3-8B通过埃洛评分机制筛选最优提示。整个过程无需额外训练数据。在三个领域、两种评估者类型及六项标准下，MK2在自动评测中排名第一，并在36场测试中赢得25场。仅材料-化学赛道表现滞后，表明仍需更强的领域知识支撑；但结果证明，轻量级提示工程已能从专利中产出具备竞争力和商业价值的创意。
原文摘要 · Abstract (English)
The Patent-Based Idea Generation task asks systems to turn real patents into product ideas viable within three years. We propose MK2, a prompt-centric pipeline: Gemini 2.5 drafts and iteratively edits a prompt, grafting useful fragments from weaker outputs; GPT-4.1 then uses this prompt to create one idea per patent, and an Elo loop judged by Qwen3-8B selects the best prompt-all without extra training data. Across three domains, two evaluator types, and six criteria, MK2 topped the automatic leaderboard and won 25 of 36 tests. Only the materials-chemistry track lagged, indicating the need for deeper domain grounding; yet, the results show that lightweight prompt engineering has already delivered competitive, commercially relevant ideation from patents.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。