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.
Figures & tables
Figure 1: Distribution of cases by number of distinct heir categories in the training and test splits. Heir categories represent distinct kinship types (e.g., father, mother, siblings). For dataset analysis, multiple individuals within the same kinship group are grouped into a single category. The training and test sets share the same distributional profile, differing only in scale.
Split
Simple
ʿAwl
Radd
Total
Training
11,079
577
344
12,000
Test
456
39
5
500
Total
11535
616
349
12500
Table 1: Distribution of inheritance cases by legal complexity
Figure 2: Cumulative pipeline success rates across stages (Step1–Step4) for each model.
Validation
Test
Model
Sheirs
Sshares
Snorm
MIR-E
Sheirs
Sshares
Snorm
MIR-E
C-ML-R
Gemini-2.5-flash
0.965
0.899
0.892
0.918
0.945
0.929
0.894
0.901
OS-ML-R
Qwen3-32B
0.761
0.439
0.237
0.438
0.690
0.446
0.265
0.437
OS-ML
GPT-oss-120B
0.692
0.311
0.273
0.389
0.693
0.327
0.271
0.391
OS-ML
Llama-3.3
0.676
0.321
0.148
0.345
0.648
0.403
0.215
0.390
OS-A
Fanar-Sadiq
0.655
0.443
0.204
0.393
0.621
0.367
0.204
0.368
Table 2: (C: commercial, OS: open-source, ML: multilingual, A: Arabic, R: reasoning.) Component-wise MIR-E scores on the validation (200 cases) and test (500 cases) sets. Validation results are used for model selection, while all comparisons and conclusions are based on the held-out test set.
Model
FB
FE
ME
Total_err
Gemini-2.5-flash
77
66
0
143
Fanar-Sadiq
285
456
0
741
GPT-OSS-120B
415
327
1
743
LLaMA-3.3-70B
287
619
1
907
Fanar-C-2-27B
168
937
0
1105
Total
1232
2405
2
3639
Table 3: Structural heir-identification errors on the test corpus. FB: false blocking; FE: false eligibility; ME: count mismatch errors; Total_err: total errors.
Figure 3: Blocking errors by genealogical heir level across models. For each model, the left bar shows false blocking (FB: wrongly blocked eligible heirs) and the right bar shows false eligibility (FE: added eligible heirs). Colors indicate heir levels (1–8), as defined in A .
Model
#(Sh=1)
#(Ss∣Sh=1)
Gemini-2.5-flash
390
0.977
Fanar-Sadiq
64
0.727
Llama-3.3-70b
74
0.612
Fanar-C-2-27B
44
0.542
GPT-oss-120b
108
0.526
Table 4: Share-allocation score conditioned on correct heir identification.
Model
#(Sh=1,Ss=1)
#(Sa=1∣Sh=1,Ss=1)
#(Sa=0∣Sh=1,Ss=1) (%)
Gemini-2.5-flash
373
372
1 (0.27)
Qwen3-32B
59
55
4 (6.78)
Fanar-Sadiq
36
33
3 (8.33)
GPT-oss-120b
26
20
6 (23.08)
Llama-3.3-70b
26
25
1 (3.85)
Fanar-C-2-27B
12
11
1 (8.33)
Table 5: Adjustment accuracy conditioned on correct previous steps (heir identification Sh=1 and share assignment Ss=1 ). We report the subset size #(Sh=1,Ss=1) and the number of correct/incorrect adjustment outcomes ( Sa=1/0 ) within this subset.
Appendix figures & tables2 assets
Supplementary material from the paper’s appendix.
Appendix
Arabic heir
English gloss
Arabic heir
English gloss
أم
Mother
أخ شقيق
Full brother
زوج
Husband
أخت شقيقة
Full sister
ابن
Son
أخ لأب
Paternal half-brother (same father)
أب
Father
أخت لأب
Paternal half-sister (same father)
زوجة
Wife
أخ لأم
Maternal half-brother (same mother)
بنت
Daughter
أخت لأم
Maternal half-sister (same mother)
Appendix
Table 6: Arabic–English glosses for the 36 heir categories used in our corpus.
Figure 4: Frequency distribution of all heir types in the corpus, ordered by kinship proximity.
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.
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 +2
This paper presents a comprehensive overview of the QIAS 2026 shared task, organized as part of the OSACT7 Workshop and co-located with LREC 2026. The shared task was designed to evaluate the ability of large language models to perform complex reasoning in the religious and legal domain of Islamic inheritance. Unlike conventional question-answering benchmarks, QIAS 2026 focuses on end-to-end reasoning from natural language cases, requiring systems to perform the full inheritance calculation process, from identifying the eligible heirs to assigning the correct share to each beneficiary. To support this evaluation, the task was based on the MAWARITH benchmark, a dataset of 12,500 Arabic inheritance cases annotated with intermediate reasoning steps and final answers. System submissions were evaluated using MIR-E, a multi-step metric that measures performance across the main stages of inheritance reasoning. A total of 16 teams participated in the shared task, investigating a range of approaches, including prompting-based methods, retrieval-augmented generation, and fine-tuning strategies. The results show that Islamic inheritance remains a highly challenging benchmark for current language models, especially in stages that require precise legal interpretation and structured numerical reasoning. This overview summarizes the task design, dataset, evaluation framework, participating systems, and main results.
This paper presents the participation of team PSL in the QIAS 2026 Shared Task on Arabic Islamic inheritance reasoning. The task evaluates the ability of large language models to solve inheritance cases that require legal interpretation, multi-step reasoning, and precise numerical computation. We compare commercial and open-source models under a unified prompting strategy to assess their effectiveness in structured legal reasoning with minimal task-specific adaptation. \ Our results show a clear gap in reliability between the two model families. Commercial models demonstrate stronger performance in identifying eligible heirs, applying exclusion rules, and maintaining consistency across reasoning steps. In contrast, open-source models exhibit greater instability, particularly in cases involving dependent legal decisions and fractional share adjustments. The best performance is achieved by Gemini 2.5 Flash, with an MRE of 0.989.
Mohammed Amine Mouhoub, Chahinez Bouchekif
Paris Dauphine University · University of Abou Bekr Belkaïd