PhD Candidate · Artificial Intelligence

Amal Saad Alshehri

PhD candidate in the Department of Computer Science at Durham University, working on Legal AI, natural language processing, legal information retrieval, reranking, and legal reasoning.

My research focuses on building and evaluating AI systems for legal text. I am particularly interested in retrieval and reranking for statutes and case law, retrieval-augmented prompting, and the careful evaluation of large language models in legal information processing tasks.

Email Durham profile ORCID LinkedIn GitHub

Amal Saad Alshehri — Legal AI researcher working on retrieval, reranking, and evaluation for legal text.

Research interests

My work sits at the intersection of artificial intelligence, natural language processing, and legal information systems. I study how models retrieve legal materials, rank evidence, and make entailment decisions under the constraints of legal language.

Legal information retrieval for statutes, case law, and legal paragraphs.
Neural and feature-based reranking for long legal documents.
Retrieval-augmented prompting and LLM evaluation for legal reasoning.
Open-weight models, reproducible legal NLP systems, and benchmark-based evaluation.
Trustworthy Legal AI, including grounding, error analysis, and transparent reporting.

Publications

Neural reranking for UK statutory retrieval: Provision-level evaluation and an open distilled model

Artificial Intelligence and Law, 2026

This work introduces UK-STATUTECORPUS and Distilled-Voyage-ModernBERT for provision-level statutory retrieval.

Paper Code and data

Cross-Architecture LLM Ensembles, Feature-Based Reranking and Retrieval-Augmented Prompting for Legal Information Processing

COLIEE 2026 workshop paper

A unified empirical study of Team DU systems across legal case retrieval, case entailment, statute retrieval and entailment, and legal judgment prediction.

COLIEE 2026

Team DU participated in all five COLIEE 2026 tasks. The strongest official result was in Task 4, where DU1 and DU2 achieved first place in statute entailment.

Task	Result	Description
Task 1	F1 = 0.314; rank 11/54 submissions	Legal case retrieval.
Task 2	Official F1 = 0.343; post-competition F1 = 0.555	Legal case entailment.
Task 3	Official accuracy = 79.3%; post-competition accuracy = 91.5%	Statute retrieval and entailment.
Task 4	Accuracy = 96.3%; rank 1/33 submissions	Official first-place result in statute entailment.
Pilot Task	TP accuracy = 73.1%; RE F1 = 68.2%	Unofficial submission for Japanese tort judgment prediction.

COLIEE results overview Task 4 results

Academic service

Reviewer, Expert Systems with Applications

Elsevier · October 2025

Completed one peer review for Expert Systems with Applications, an Elsevier journal focused on expert and intelligent systems applied in industry, government, and universities worldwide.

View certificate

Certificate of reviewing for Expert Systems with Applications

Reviewing certificate

Presentations

UKAIS 2026

University of Sheffield, April 2026

Presented “Enhancing Trust in Legal AI: Optimising Span-Level Retrieval Architectures on LegalBench-RAG.”

View slides Download slides

Contact

I am happy to connect with researchers and collaborators interested in Legal AI, NLP, information retrieval, reranking, and trustworthy AI systems for legal text.

Email: amal.alshehri@durham.ac.uk

Durham: University profile

ORCID: 0000-0001-6903-6720

LinkedIn: amalsaad100