用集成学习在法律文档中高效找关键信息，不依赖深度学习。
Finding Needles in Emb(a)dding Haystacks: Legal Document Retrieval via Bagging and SVR Ensembles
- 将检索任务拆解为多个二分类子任务，结合袋装法与支持向量回归
- 在GerDaLIR数据集上召回率达0.849，优于基线方法
- 适合无需训练大模型的法律文本检索场景
我们提出一种基于支持向量回归（SVR）集成、自举聚合（bagging）和嵌入空间的检索方法，应用于德国法律信息检索数据集（GerDaLIR）。通过将检索任务建模为多个二分类的“针在 haystack”子任务，我们的投票集成方法实现了0.849的召回率，优于基线（0.803 | 0.829），且无需训练或微调任何深度学习模型。该方法具有进一步提升潜力，尤其可通过优化编码模型和超参数实现改进。
原文摘要 · Abstract (English)
We introduce a retrieval approach leveraging Support Vector Regression (SVR) ensembles, bootstrap aggregation (bagging), and embedding spaces on the German Dataset for Legal Information Retrieval (GerDaLIR). By conceptualizing the retrieval task in terms of multiple binary needle-in-a-haystack subtasks, we show improved recall over the baselines (0.849 > 0.803 | 0.829) using our voting ensemble, suggesting promising initial results, without training or fine-tuning any deep learning models. Our approach holds potential for further enhancement, particularly through refining the encoding models and optimizing hyperparameters.
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。