发表机构
VNU University of Engineering and Technology; Center for Juris-Informatics, ROIS-DS College of Engineering & Computer Science, VinUniversity(越南国立工程技术大学; 越南Vin大学工程与计算机科学学院法律信息学中心)
机构由 AI 辅助整理,请以论文原文为准。AI 中文总结
NOWJ团队参与COLIEE 2026竞赛五项任务,针对各任务提出不同方法。如任务1的四阶段管道,任务2的多种方法结合,任务3的检索增强框架等,通过这些方法在法律检索、推理等方面取得相应成果。
AI 中文摘要
本文介绍了NOWJ团队参与COLIEE 2026竞赛所有五项任务的方法和结果。对于任务1(法律案例检索),提出了一个四阶段管道,包括候选过滤、使用互补嵌入模型的密集检索、通过微调生成重排器的交叉编码器重排和基于MLP的成对分类,以及自适应逐查询截止预测。对于任务2(法律案例蕴含),将BM25过滤、基于T5的重排和基于大语言模型的蕴含验证与共识集成相结合。对于任务3(成文法检索和蕴含),采用了一个检索增强生成框架,包括密集检索、基于注意力的重排和少样本提示大语言模型推理。对于任务4(法律文本蕴含),引入了一个动态路由管道,用于对查询难度进行分类,并将案例分配给平衡少样本求解器或结构化零样本思维链求解器。对于试点任务(法律判决预测),将分层变压器与CRF层相结合,并进行论证关系挖掘和概率论证图推理。
英文摘要
This paper presents the methodologies and results of the NOWJ team's participation across all five tasks of the COLIEE 2026 competition. For Task 1 (Legal Case Retrieval), we propose a four-stage pipeline comprising candidate filtering, dense retrieval with complementary embedding models, cross-encoder reranking via fine-tuned generative rerankers and MLP-based pairwise classification, and adaptive per-query cutoff prediction. For Task 2 (Legal Case Entailment), we combine BM25 filtering, T5-based reranking, and LLM-based entailment verification with consensus ensemble. For Task 3 (Statute Law Retrieval and Entailment), we adopt a retrieval-augmented generation framework with dense retrieval, attention-based reranking, and few-shot-prompted LLM reasoning. For Task 4 (Legal Textual Entailment), we introduce a dynamic routing pipeline that classifies query difficulty and dispatches cases to either a balanced few-shot solver or a structured zero-shot chain-of-thought solver. For the Pilot Task (Legal Judgment Prediction), we combine hierarchical transformers with CRF layers, argument relation mining, and probabilistic argumentation graph reasoning.
CommentsPresented at COLIEE 2026