CVPD at QIAS 2026: RAG-Guided LLM Reasoning for Al-Mawarith Share Computation and Heir Allocation

Wassim Swaileh; Mohammed-En-Nadhir Zighem; Hichem Telli; Salah Eddine Bekhouche; Abdellah Zakaria Sellam; Fadi Dornaika; Dimitrios Kotzinos

cs.CLMar 25, 2026

CVPD at QIAS 2026: RAG-Guided LLM Reasoning for Al-Mawarith Share Computation and Heir Allocation

Authors: Wassim Swaileh, Mohammed-En-Nadhir Zighem, Hichem Telli, Salah Eddine Bekhouche, Abdellah Zakaria Sellam, Fadi Dornaika, Dimitrios Kotzinos

Organizations: ETIS (UMR 8051), CY Cergy Paris Univ., ENSEA, CNRS, France · Computer Engineering Dept, Sana’a Community College, Yemen · VSC Laboratory, Department of Electronics and Automation, University of Biskra, Algeria · Dept. of CS & AI, Univ. of the Basque Country (UPV/EHU), Spain · CNR-ISASI "E. Caianiello", Lecce, Italy

Abstract

Islamic inheritance (Ilm al-Mawarith) is a multi-stage legal reasoning task requiring the identification of eligible heirs, resolution of blocking rules (hajb), assignment of fixed and residual shares, handling of adjustments such as awl and radd, and generation of a consistent final distribution. The task is further complicated by variations across legal schools and civil-law codifications, requiring models to operate under explicit legal configurations. We present a retrieval-augmented generation (RAG) pipeline for this setting, combining rule-grounded synthetic data generation, hybrid retrieval (dense and BM25) with cross-encoder reranking, and schema-constrained output validation. A symbolic inheritance calculator is used to generate a large high-quality synthetic corpus with full intermediate reasoning traces, ensuring legal and numerical consistency. The proposed system achieves a MIR-E score of 0.935 and ranks first on the official QIAS 2026 blind-test leaderboard. Results demonstrate that retrieval-grounded, schema-aware generation significantly improves reliability in high-precision Arabic legal reasoning tasks.

Figures & tables

Figure 1: Overview of the proposed RAG pipeline for QIAS.

Category	Train	Dev	Diff
Simple	5,531 (93.8%)	125 (62.5%)	-31.3%
‘Awl	288 (4.9%)	4 (2.0%)	-2.9%
Radd	81 (1.4%)	71 (35.5%)	+34.1%
Total	5,900	200	–

Table 1: QIAS 2026 dataset distribution showing train–dev imbalance.

Model	MIR-E	Missing
Qwen 3.5 9B	0.347	19
Qwen2.5-7B	0.243	0
Command-R7B	0.225	32
Llama3.2-1B	0.109	80
Gemma3-12B	0.010	193

Table 2: Off-the-shelf benchmarking on the 200-case QIAS dev set.

Rank	Team	MIR-E
1	CVPD (ours)	0.935
2	Simplicity Almansour (2026)	0.931
3	KMS Alkhamis (2026)	0.916
4	QU-NLP *	0.907
5	PSL Mouhoub (2026)	0.898
6	grkurdi *	0.826

Table 3: Official QIAS 2026 blind-test leaderboard. Teams marked with * did not have published system descriptions at the time of writing.

Source	Combined	Semantic	Keyword
PDF-only	0.718	0.879	0.235
PDF+Web	0.591	0.746	0.229
Web-only	0.463	0.593	0.269

Table 4: Source ablation on the 200-question QIAS dev set.

Explore similar work

MAWARITH: A Dataset and Benchmark for Legal Inheritance Reasoning with LLMs
Mar 8, 2026Abdessalam Bouchekif, Shahd Gaben, Samer Rashwani +5Legal Reasoning LLM Evaluation
Hamad Bin Khalifa University, Qatar · Nazarbayev University, Kazakhstan
QIAS 2026: Overview of the Shared Task on Islamic Inheritance Reasoning
Jun 11, 2026Abdessalam Bouchekif, Somaya Eltanbouly, Samer Rashwani +5LLM Evaluation Legal Reasoning
Hamad bin Khalifa University, Qatar · Nazarbayev University, Kazakhstan
QU-NLP at QIAS 2026: Multi-Stage QLoRA Fine-Tuning for Arabic Islamic Inheritance Reasoning
Mar 29, 2026Mohammad AL-SmadiLegal Reasoning Arabic NLP
Qatar University Doha, Qatar

Mar 8, 2026J/K move · Enter open · S save

MAWARITH: A Dataset and Benchmark for Legal Inheritance Reasoning with LLMs

Abdessalam Bouchekif, Shahd Gaben, Samer Rashwani, Somaya Eltanbouly+4

Hamad Bin Khalifa University, Qatar · Nazarbayev University, Kazakhstan

Legal Reasoning LLM Evaluation Arabic NLP

Open paper AskPDF arXiv

Islamic inheritance law is challenging for large language models because solving inheritance cases requires complex, structured, multi-step reasoning and the correct application of juristic rules to compute heirs' shares. We introduce MAWARITH, a large-scale annotated dataset of 12,500 Arabic inheritance cases for training and evaluating models on the full reasoning chain: (i) identifying eligible heirs, (ii) applying blocking (\d{h}ajb) and allocation rules, and (iii) computing exact inheritance shares. To the best of our knowledge, MAWARITH is the first Arabic corpus and benchmark designed for end-to-end Islamic inheritance reasoning. Unlike prior datasets that restrict inheritance case solving to multiple-choice questions, MAWARITH supports the full reasoning chain and provides step-by-step solutions with justifications grounded in classical juristic sources and established inheritance rules, as well as exact share calculations. This enables models to learn how to generate detailed, step-by-step responses to user queries that reflect real-world Islamic inheritance cases. To evaluate models beyond final-answer accuracy, we propose MIR-E (Mawarith Inheritance Reasoning Evaluation), a weighted multi-stage metric that scores key reasoning stages and captures error propagation across the pipeline. We evaluate six large language models in a zero-shot setting. A commercial model achieves about 90%, whereas all evaluated open-source models remain below 50%. 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 such as \textquotesingle awl and radd. The MAWARITH dataset is publicly available at https://gitlab.com/nlpresearcher/mawarith.

Abstract

Explore similar work

MAWARITH: A Dataset and Benchmark for Legal Inheritance Reasoning with LLMs

QIAS 2026: Overview of the Shared Task on Islamic Inheritance Reasoning

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