提升专利权利要求生成质量，跨司法管辖区表现更稳定。
Adaptive Multi-Stage Patent Claim Generation with Unified Quality Assessment
- 分三阶段生成：关系建模、领域自适应生成、统一质量评估。
- 在多个数据集上比GPT-4o高7.6点ROUGE-L，BERTScore提升8.3%。
- 适合需要高可靠专利自动化工具的法律与科技机构。
现有专利权利要求生成系统存在三大局限：跨司法管辖区泛化能力差、权利要求与现有技术间语义关系建模不足、质量评估不可靠。本文提出一种三阶段框架，通过关系感知的相似性分析、领域自适应的权利要求生成以及统一的质量评估来解决这些问题。方法采用八头注意力机制显式建模语义关系，结合课程学习与动态LoRA适配器选择，在五个专利领域中实现自适应生成，并通过交叉注意力机制在评估维度间建立关联，实现全面质量评估。在USPTO HUPD、EPO专利集合及Patent-CE基准上的实验表明，相比GPT-4o提升7.6点ROUGE-L，Llama-3.1-8B的BERTScore提升8.3%，与人工专家相关性达0.847（基线为0.623）。跨司法管辖区性能保留率达89.4%（基线76.2%），为自动化专利审查流程提供完整解决方案。
原文摘要 · Abstract (English)
Current patent claim generation systems face three fundamental limitations: poor cross-jurisdictional generalization, inadequate semantic relationship modeling between claims and prior art, and unreliable quality assessment. We introduce a novel three-stage framework that addresses these challenges through relationship-aware similarity analysis, domain-adaptive claim generation, and unified quality assessment. Our approach employs multi-head attention with eight specialized heads for explicit relationship modeling, integrates curriculum learning with dynamic LoRA adapter selection across five patent domains, and implements cross-attention mechanisms between evaluation aspects for comprehensive quality assessment. Extensive experiments on USPTO HUPD dataset, EPO patent collections, and Patent-CE benchmark demonstrate substantial improvements: 7.6-point ROUGE-L gain over GPT-4o, 8.3\% BERTScore enhancement over Llama-3.1-8B, and 0.847 correlation with human experts compared to 0.623 for separate evaluation models. Our method maintains 89.4\% cross-jurisdictional performance retention versus 76.2\% for baselines, establishing a comprehensive solution for automated patent prosecution workflows.
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。