QU-NLP at QIAS 2026: Multi-Stage QLoRA Fine-Tuning for Arabic Islamic Inheritance Reasoning

• URL: http://arxiv.org/abs/2604.16396v1
• Date: Sun, 29 Mar 2026 09:11:17 GMT
• Title: QU-NLP at QIAS 2026: Multi-Stage QLoRA Fine-Tuning for Arabic Islamic Inheritance Reasoning
• Abstract summary: We present QU-NLP's submission to the QIAS 2026 shared task on Arabic Islamic inheritance reasoning.<n>Our approach employs a multi-stage Quantized Low-Rank Adaptation strategy on Qwen3-4B.<n>Using 4-bit NF4 quantization with rank-128 LoRA, our model achieves 90% MIR-E (Mawarith Inheritance Reasoning Evaluation) score on the test set.
• Score: 1.0152838128195467
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: 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.

Related papers

• Which Models Perform Better in Inheritance Reasoning? [0.0]
  This paper presents the participation of team PSL in the QIAS 2026 Shared Task on Arabic Islamic inheritance reasoning.<n>The task evaluates the ability of large language models to solve inheritance cases that require legal interpretation, multi-step reasoning, and precise numerical computation.<n>We compare textitcommercial and textitopen-source models under a unified prompting strategy to assess their effectiveness in structured legal reasoning with minimal task-specific adaptation.
  arXiv  Detail & Related papers  (2026-06-11T15:23:32Z)
• LLMEval-Logic: A Solver-Verified Chinese Benchmark for Logical Reasoning of LLMs with Adversarial Hardening [69.1037790901185]
  We present LLMEval-Logic, a Chinese logical reasoning benchmark built from realistic situational scenarios.<n>Its pipeline forward-authors and expert-audits natural-language items together with their reference formalizations, verifies annotated answers with Z3, constructs expert rubrics for natural-to-formal grading, and hardens selected items through a closed-loop adversarial workflow.
  arXiv  Detail & Related papers  (2026-05-19T09:40:29Z)
• DeonticBench: A Benchmark for Reasoning over Rules [52.69517904415795]
  DEONTICBENCH is a benchmark of 6,232 tasks across U.S. federal taxes, airline baggage policies, U.S. immigration administration, and U.S. state housing law.<n>It is a benchmark for studying context-grounded rule reasoning in real-world domains under both symbolic and non-symbolic settings.
  arXiv  Detail & Related papers  (2026-04-06T05:41:02Z)
• MAWARITH: A Dataset and Benchmark for Legal Inheritance Reasoning with LLMs [0.16060719742433224]
  We introduce MAWARITH, a large-scale annotated dataset of 12,500 Arabic inheritance cases for training and evaluating models.<n>Unlike prior datasets that restrict inheritance case solving to multiple-choice questions, MAWARITH supports the full reasoning chain.<n>Our error analysis identifies recurring failure patterns, including scenario misinterpretation, errors in heir identification, errors in share allocation, and missing or incorrect application of key inheritance rules.
  arXiv  Detail & Related papers  (2026-03-08T08:54:33Z)
• Assessing Large Language Models on Islamic Legal Reasoning: Evidence from Inheritance Law Evaluation [0.17592522344393483]
  o3 and Gemini 2.5 achieved accuracies above 90%, whereas ALLaM, Fanar, LLaMA, and Mistral scored below 50%.<n>We conduct a detailed error analysis to identify recurring failure patterns across models.<n>Our findings highlight limitations in handling structured legal reasoning and suggest directions for improving performance in Islamic legal reasoning.
  arXiv  Detail & Related papers  (2025-09-01T03:08:10Z)
• QU-NLP at QIAS 2025 Shared Task: A Two-Phase LLM Fine-Tuning and Retrieval-Augmented Generation Approach for Islamic Inheritance Reasoning [1.0152838128195467]
  We fine-tuned the Fanar-1-9B causal language model using Low-Rank Adaptation (LoRA) and integrated it into a Retrieval-Augmented Generation pipeline.<n>Our system achieves an accuracy of 0.858 in the final test, outperforming other competitive models such as, GPT 4.5, LLaMA, Fanar, Mistral and ALLaM evaluated with zero-shot prompting.
  arXiv  Detail & Related papers  (2025-08-20T10:29:55Z)
• Benchmarking the Legal Reasoning of LLMs in Arabic Islamic Inheritance Cases [1.3521447196536418]
  Islamic inheritance domain holds significant importance for Muslims to ensure fair distribution of shares between heirs.<n>Recent advancements in Large Language Models (LLMs) have sparked interest in their potential to assist with complex legal reasoning tasks.<n>This study evaluates the reasoning capabilities of state-of-the-art LLMs to interpret and apply Islamic inheritance laws.
  arXiv  Detail & Related papers  (2025-08-13T10:37:58Z)
• DARA: Decomposition-Alignment-Reasoning Autonomous Language Agent for Question Answering over Knowledge Graphs [70.54226917774933]
  We propose the DecompositionAlignment-Reasoning Agent (DARA) framework. DARA effectively parses questions into formal queries through a dual mechanism. We show that DARA attains performance comparable to state-of-the-art enumerating-and-ranking-based methods for KGQA.
  arXiv  Detail & Related papers  (2024-06-11T09:09:37Z)
• Lory: Fully Differentiable Mixture-of-Experts for Autoregressive Language Model Pre-training [73.90260246781435]
  We present Lory, the first approach that scales such architectures to autoregressive language model pre-training. We show significant performance gains over parameter-matched dense models on both perplexity and a variety of downstream tasks. Despite segment-level routing, Lory models achieve competitive performance compared to state-of-the-art MoE models with token-level routing.
  arXiv  Detail & Related papers  (2024-05-06T03:06:33Z)
• Text Classification via Large Language Models [63.1874290788797]
  We introduce Clue And Reasoning Prompting (CARP) to address complex linguistic phenomena involved in text classification. Remarkably, CARP yields new SOTA performances on 4 out of 5 widely-used text-classification benchmarks. More importantly, we find that CARP delivers impressive abilities on low-resource and domain-adaptation setups.
  arXiv  Detail & Related papers  (2023-05-15T06:24:45Z)
• Lila: A Unified Benchmark for Mathematical Reasoning [59.97570380432861]
  LILA is a unified mathematical reasoning benchmark consisting of 23 diverse tasks along four dimensions. We construct our benchmark by extending 20 datasets benchmark by collecting task instructions and solutions in the form of Python programs. We introduce BHASKARA, a general-purpose mathematical reasoning model trained on LILA.
  arXiv  Detail & Related papers  (2022-10-31T17:41:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.