用检索增强方法解决水电许可文件中环保措施分类标签稀缺问题
Retrieval-Augmented Classification of Environmental Mitigations in Hydropower Licensing Documents
- 引入检索增强生成机制，通过类别定义实现零样本分类
- 混合模型在全部标签类别上达到0.524的微平均F1，优于纯BERT或纯RAG
- 适合需要处理低资源、多标签环境评估任务的研究与实务人员
识别并分类联邦能源监管委员会水电许可文件中的环境缓解义务是一项耗时且需深厚领域知识的任务。我们将其建模为一个包含135个类别的多标签分类问题，面临严重标签稀缺：135个类别中有40个无训练样本，26个少于5个。基于BERT的监督管道虽在常见类别上表现良好，但在未见类别上F1为零，无论采用何种增强策略。我们提出一种检索增强生成（RAG）管道，通过检索类别定义来引导分类，实现全标签空间的零样本泛化。进一步设计混合系统，结合微调后的BERT检测与RAG分类，兼顾高召回率与零样本覆盖能力。在2017年全部许可文档（5,860段落，135类别）上评估，混合模型取得0.524的微平均F1，优于仅用BERT（0.477）和仅用RAG（0.416）的模型，且在所有训练支持度分桶中均表现更优。
原文摘要 · Abstract (English)
Identifying and classifying environmental mitigation obligations in Federal Energy Regulatory Commission hydropower licensing documents is a labor-intensive task requiring deep domain expertise. We formulate this as a multi-label classification problem over a structured 135-category taxonomy and address the central challenge of severe label scarcity: 40 of 135 categories have no training examples, and 26 have fewer than five. A supervised Bidirectional Encoder Representations from Transformers (BERT)-based pipeline, while effective on well-represented categories, achieves F1 of zero on unseen classes regardless of augmentation strategy. We introduce a Retrieval-Augmented Generation (RAG) pipeline that conditions classification on retrieved category definitions, enabling zero-shot generalization across the full label space. We further propose a hybrid system that combines BERT detection with RAG classification, exploiting the high recall of fine-tuned detection and the zero-shot coverage of retrieval-augmented reasoning. Evaluated on the full set of 2017 license documents (5,860 paragraphs, 135 categories), the hybrid achieves a Micro F1 of 0.524, outperforming the BERT-only pipeline (0.477) and the RAG-only pipeline (0.416) across all training-support buckets.
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。