Skip to content
Review Open access

Domain Adaptation of Foundation Language Models for Legal Document Generation: A Systematic Review and Implications for Analogous Regulated Institutional Contexts

Oct 2026 · Revista de Ciência da Computação · Vol 8, pp. e19945 · 0 citations · 49 references

TL;DR

The findings indicate that the institutional adoption of foundation language models for legal document generation should be treated as a problem of architecture, evaluation and governance, rather than only textual performance.

Abstract

The adaptation of foundation language models for legal document generation has become a relevant problem in Legal Natural Language Processing, particularly because generated texts must preserve factuality, normative coherence, traceability and control over sensitive information. This systematic literature review maps, classifies and critically analyzes primary studies on domain adaptation techniques for foundation language models applied to legal and normative-argumentative document generation. It also examines the implications of the resulting evidence for analogous regulated institutional contexts that share requirements related to reliability, privacy, auditability and institutional control. The review followed guidelines for systematic reviews in Software Engineering and the applicable items of PRISMA 2020. Searches were conducted in Scopus, Web of Science, IEEE Xplore and ACM Digital Library, covering studies published from January 2021 to December 2025 and complemented by backward snowballing. From 200 initial records, 32 primary studies were included. The corpus is predominantly concentrated in the legal domain, with only one study directly addressing document production in Public Security. The results show a recurrent use of hybrid adaptation strategies involving parameter-efficient fine-tuning, retrieval-augmented generation and structured inference, although the literature does not establish a universally superior combination. Data engineering is marked by a trade-off between scale and control, with frequent use of semi-synthetic datasets. Traditional Natural Language Processing metrics remain useful as initial indicators, but are insufficient to assess factuality, normative validity and argumentative consistency. Ethical safeguards, privacy controls and bias mitigation metrics remain underdeveloped. The findings indicate that the institutional adoption of foundation language models for legal document generation should be treated as a problem of architecture, evaluation and governance, rather than only textual performance. These conclusions are directly grounded in the Legal NLP literature, while their transfer to analogous regulated institutional contexts should be understood as a functionally motivated implication requiring domain-specific empirical validation.

Read PDF

Similar papers

Review Open access Sep 2026

Detecting Contradictions in Bilingual Legislation: A Systematic Review and Conceptual Architecture

A conceptual four-layer architecture combining legislative text normalization, knowledge-graph representation, hybrid semantic reasoning and explainable human oversight is proposed, and research priorities for reliable and transparent AI-assisted legislative analysis are identified.

Maxatbek Satymbekov, A. Yeleussinov, Z. Buribayev et al. · 0 citations
Open access Sep 2026

Development and systematization of procedural documents in the context of their analysis by artificial intelligence tools in the process of citizens’ access to justice in the Republic of Kazakhstan

This study systematically analyzes procedural documents of pre-trial proceedings, judicial acts, and legal instruments drafted by citizens and participants in legal proceedings to protect their rights and freedoms acrossvarious branches of law to evaluate the feasibility of adapting artificial intelligence tools tobuil...

S. Bachurin, T. Khanov, Y. Tileubergenov et al. · 0 citations
Open access Oct 2026

Building a Multilingual Legal Risk Database and ClauseRisk Visual Analytics System for Contracts and Online Service Agreements

Documents containing binding legal terms and risk obligations remain largely unread, particularly in multilingual contexts where regulatory systems vary across jurisdictions. This paper presents a multilingual Retrieval-Augmented Generation (RAG) system designed to identify potentially risky clauses across Arabic, Engl...

Mohammed El Seddik Roukbi, Samiha Fadloun, Souham Meshoul et al. · 0 citations
Conference

Exploring Large Language Models as Decision Support Tools: A Proof of Concept for Procedure-Related Information Retrieval

Policy and procedural documentation are essential for the effective operation of government agencies. However, the vast volume of these documents often obscures critical information within hundreds of pages of unrelated content. Recent advancements in large language models (LLMs) enhance text search and response genera...

Nathaniel Shepherd, Thomas Berg, Xue-Ping Li · 0 citations
Preprint Oct 2026

TenetBench: A Clause-Anchored Russian Corpus for Evaluating Language Models against Professional Codes of Ethics

Codes of professional ethics guide and restrict specialists, who increasingly work with assistants built on large language models. Such an assistant can breach the user's code, or help the user breach it, without the user noticing. Evaluations reporting broad categories of harm cannot show how often this happens or whi...

Yegor Anichkov, V. Vinogradov, Andrey Furov et al. · 0 citations
Open access Oct 2026

RAG Based Legal Document Assistant with Verifiable Citations and Multilingual Support

The increasing availability of legal information in digital form has created a need for intelligent systems that can provide reliable and understandable access to legal documents. Conventional language models may generate fluent responses but can produce unsupported information, particularly when answering complex lega...

Gayatri Sampat Gavhane, Prof. G. R. Sarode, Dr. H. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.