# CVPD at QIAS 2025 Shared Task: An Efficient Encoder-Based Approach for Islamic Inheritance Reasoning

## TL;DR

This paper introduces a lightweight, encoder-based AI system for Islamic inheritance reasoning that balances accuracy with efficiency and privacy, enabling on-device deployment for high-stakes legal questions.

## Contribution

It presents a specialized Arabic text encoder with Attentive Relevance Scoring for inheritance questions, demonstrating competitive accuracy with significantly lower resource requirements.

## Key findings

- MARBERT-based model achieves 69.87% accuracy
- Large LLMs reach up to 87.6% accuracy but are resource-intensive
- The approach enables fast, privacy-preserving on-device inference

## Abstract

Islamic inheritance law (Ilm al-Mawarith) requires precise identification of heirs and calculation of shares, which poses a challenge for AI. In this paper, we present a lightweight framework for solving multiple-choice inheritance questions using a specialised Arabic text encoder and Attentive Relevance Scoring (ARS). The system ranks answer options according to semantic relevance, and enables fast, on-device inference without generative reasoning. We evaluate Arabic encoders (MARBERT, ArabicBERT, AraBERT) and compare them with API-based LLMs (Gemini, DeepSeek) on the QIAS 2025 dataset. While large models achieve an accuracy of up to 87.6%, they require more resources and are context-dependent. Our MARBERT-based approach achieves 69.87% accuracy, presenting a compelling case for efficiency, on-device deployability, and privacy. While this is lower than the 87.6% achieved by the best-performing LLM, our work quantifies a critical trade-off between the peak performance of large models and the practical advantages of smaller, specialized systems in high-stakes domains.