Dataset
patent-drafting-bench
patent-drafting-bench 中文 | English patent-drafting-bench 是一个评估大模型及智能体专利撰写能力的数据集,包含 60 个测试任务。模型需要依据技术交底书,按指定法域和语言撰写权利要求书、说明书及必要附图,最终交付一份 Word(.docx)文档。 评测关注模型能否将交底中的技术方案转化为完整、合理且有依据的专利申请文本,包括权利要求的完整性、清楚性、保护范围和从属布局,说明书的支持与公开,以及是否编造技术内容。每个任务包含通用评分标准与案例特异评分标准,可用于逐项人工审阅或 LLM judge 评分。 目录结构 patent-drafting-bench/ ├── README.md ├── README.en.md ├── data/ │ └── test.jsonl └── asset/ ├── PD-0005_input.docx
Data preview
Dataset card
patent-drafting-bench is listed on Hugging Face by PatSnap.
The recorded metadata describes free-text content, stored in JSON format and estimated at under a thousand records. The listing has drawn 144 downloads and 0 likes on Hugging Face.
The data
Organized as free-text content, in Chinese.
The dataset ships in JSON format and is sized at under a thousand records.
The source tags it with patent, patent-drafting and benchmark.
Documentation & references 2
Discussion
Questions or observations about this dataset? Join the conversation on X.
No comments yet. Start the conversation on X.