用小模型实现阿拉伯伊斯兰继承法的精准推理，仅需少量算力。
QU-NLP at QIAS 2026: Multi-Stage QLoRA Fine-Tuning for Arabic Islamic Inheritance Reasoning
- 分阶段微调：先用3166条教法判例学术语，再用1.2万案例练结构化输出
- 4比特量化+低秩适配器，测试集得分90% MIR-E，媲美大模型
- 适合资源有限但需高精度法律推理的场景
伊斯兰继承法（ilm al-mawarıth）对大语言模型的结构化推理能力构成严峻挑战，需多步法律分析、基于规则的决策及精确的分数计算。我们提交了QU-NLP团队参与2026年QIAS共享任务的方案。方法基于Qwen3-4B模型，采用分阶段量化低秩适配（QLoRA）微调：首先在3,166条伊斯兰教法判例上进行领域适应，学习继承术语与法学推理模式；随后在12,000个结构化继承案例上进行任务特定训练，优化JSON格式输出生成。使用4位NF4量化与秩为128的LoRA适配器，模型在测试集上达到90% MIR-E（Mawarith Inheritance Reasoning Evaluation）得分，展现出优异性能，同时仅需极少计算资源。结果表明，结合领域预适应与结构化输出训练，小模型可有效完成复杂法律推理任务，性能优于如Gemini-2.5-flash等商用系统。
原文摘要 · Abstract (English)
Islamic inheritance law (ilm al-mawarıth) presents a challenging domain for evaluating large language models' structured reasoning capabilities, requiring multi-step legal analysis, rule-based blocking decisions, and precise fractional calculations. We present QU-NLP's submission to the QIAS 2026 shared task on Arabic Islamic inheritance reasoning. Our approach employs a multi-stage Quantized Low-Rank Adaptation (QLoRA) fine-tuning strategy on Qwen3-4B: (1) domain adaptation on 3,166 Islamic fatwa records to acquire inheritance terminology and jurisprudential reasoning patterns, followed by (2) task-specific training on 12,000 structured inheritance cases to optimize JSON-formatted output generation. Using 4-bit NF4 quantization with rank-128 LoRA adapters, our model achieves 90% MIR-E (Mawarith Inheritance Reasoning Evaluation) score on the test set, demonstrating competitive performance while requiring minimal computational resources. Our results show that domain-specific pre-adaptation combined with structured output training enables small language models to perform complex legal reasoning tasks effectively comparing to commercial systems such as Gemini-2.5-flash.
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。