用智能分块和图遍历让大模型生成符合汽车法规的测试场景。
Think Inside the Chunk: RegulaRAG for Regulation-Compliant Scenario Generation using LLMs: A Case Study of UN Regulation No. 152
- 通过智能分块与图遍历增强文本，提升长篇法规理解能力。
- 在联合国第152号法规测试中，平均得分82.99，比最佳基线高43%。
- 适合自动驾驶安全验证团队，尤其关注法规合规性生成。
生成符合法规的测试场景对验证安全关键型汽车系统至关重要，但大语言模型（LLMs）难以在长篇、层级化的标准中实现有效定位。我们提出RegulaRAG，一种检索增强生成（RAG）流程，结合智能分块、基于图遍历的段落与表格参考增强，以及对增强单元的智能检索与重排序。为验证系统，我们在覆盖联合国第152号法规（AEBS）全部场景的手工标注数据集上进行评估。研究包括：(i) 三步渐进式搜索，无需全网格搜索即可识别近似最优检索参数；(ii) 与五种基线RAG系统的直接对比；(iii) 针对含干扰内容的源语料库扩展的鲁棒性压力测试。输出采用定制化惩罚评分指标评估。所有实验中，RegulaRAG取得最高平均元得分（82.99），优于次优系统43%（NoRAG: 57.94），且每查询仅需14k–25k tokens，远低于图中心基线的500k。随着法规来源数量增加，其性能保持稳定，而其他RAG系统质量与鲁棒性显著下降。
原文摘要 · Abstract (English)
Generating regulation-compliant test scenarios is essential for validating safety-critical automotive systems, yet Large Language Models (LLMs) struggle to ground outputs in long, hierarchical standards. We present RegulaRAG, a Retrieval-Augmented Generation (RAG) pipeline that couples SmartChunking, reference-aware enrichment of paragraphs and tables via graph traversal, with Smart Retrieve & Rerank over these enriched units. To test our system, we evaluate on a manually curated dataset covering all scenarios in UN Regulation No. 152 (AEBS). Our study comprises: (i) a three-step progressive search that identifies near-optimal retrieval parameters without exhaustive grid search; (ii) head-to-head comparisons against five baseline RAG systems; and (iii) a robustness stress test that scales the source corpus with distractor content. Outputs are evaluated using a customized penalized scoring metric. Across all experiments, RegulaRAG achieves the highest average Meta-Score (82.99), outperforming the next-best system by 43% (NoRAG: 57.94), while operating at 14k-25k tokens per query versus up to 500k for graphcentric baselines. It maintains strong performance, remaining stable even as the number of regulatory sources grows, whereas competing RAG systems degrade sharply in both quality and robustness.
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。