@inproceedings{nguyen-etal-2026-vilegallm,
title = "{V}i{L}egal{LM}: Language Models for {V}ietnamese Legal Text",
author = "Nguyen, Truong-Phuc and
Nguyen, Quy-Nhan and
Nguyen, Minh-Tien",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.1801/",
pages = "36136--36150",
ISBN = "979-8-89176-395-1",
abstract = "We present **ViLegalLM**, comprising **ViLegalBERT** and **ViLegalQwen**, the first suite of Vietnamese pretrained language models for legal text understanding and generation. It includes one encoder-only model (ViLegalBERT, 135M parameters) and two decoder-only models (ViLegalQwen2.5-1.5B-Base and ViLegalQwen3-1.7B-Base), all continually pretrained on a newly curated 16GB Vietnamese legal corpus, significantly larger than previous work. To mitigate data scarcity, we construct three synthetic datasets using LLM-based generation and hard negative mining for True/False QA, Multiple Choice QA, and Natural Language Inference. We establish state-of-the-art results among open-source models on four main Vietnamese legal downstream tasks spanning ten benchmarks, demonstrating that continual pretraining from base models consistently outperforms instruction-tuned adaptation. Source codes, corpus, datasets, and model checkpoints are publicly available at https://github.com/ntphuc149/ViLegalLM."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="nguyen-etal-2026-vilegallm">
<titleInfo>
<title>ViLegalLM: Language Models for Vietnamese Legal Text</title>
</titleInfo>
<name type="personal">
<namePart type="given">Truong-Phuc</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Quy-Nhan</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minh-Tien</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>We present **ViLegalLM**, comprising **ViLegalBERT** and **ViLegalQwen**, the first suite of Vietnamese pretrained language models for legal text understanding and generation. It includes one encoder-only model (ViLegalBERT, 135M parameters) and two decoder-only models (ViLegalQwen2.5-1.5B-Base and ViLegalQwen3-1.7B-Base), all continually pretrained on a newly curated 16GB Vietnamese legal corpus, significantly larger than previous work. To mitigate data scarcity, we construct three synthetic datasets using LLM-based generation and hard negative mining for True/False QA, Multiple Choice QA, and Natural Language Inference. We establish state-of-the-art results among open-source models on four main Vietnamese legal downstream tasks spanning ten benchmarks, demonstrating that continual pretraining from base models consistently outperforms instruction-tuned adaptation. Source codes, corpus, datasets, and model checkpoints are publicly available at https://github.com/ntphuc149/ViLegalLM.</abstract>
<identifier type="citekey">nguyen-etal-2026-vilegallm</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.1801/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>36136</start>
<end>36150</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ViLegalLM: Language Models for Vietnamese Legal Text
%A Nguyen, Truong-Phuc
%A Nguyen, Quy-Nhan
%A Nguyen, Minh-Tien
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F nguyen-etal-2026-vilegallm
%X We present **ViLegalLM**, comprising **ViLegalBERT** and **ViLegalQwen**, the first suite of Vietnamese pretrained language models for legal text understanding and generation. It includes one encoder-only model (ViLegalBERT, 135M parameters) and two decoder-only models (ViLegalQwen2.5-1.5B-Base and ViLegalQwen3-1.7B-Base), all continually pretrained on a newly curated 16GB Vietnamese legal corpus, significantly larger than previous work. To mitigate data scarcity, we construct three synthetic datasets using LLM-based generation and hard negative mining for True/False QA, Multiple Choice QA, and Natural Language Inference. We establish state-of-the-art results among open-source models on four main Vietnamese legal downstream tasks spanning ten benchmarks, demonstrating that continual pretraining from base models consistently outperforms instruction-tuned adaptation. Source codes, corpus, datasets, and model checkpoints are publicly available at https://github.com/ntphuc149/ViLegalLM.
%U https://aclanthology.org/2026.findings-acl.1801/
%P 36136-36150
Markdown (Informal)
[ViLegalLM: Language Models for Vietnamese Legal Text](https://aclanthology.org/2026.findings-acl.1801/) (Nguyen et al., Findings 2026)
ACL
- Truong-Phuc Nguyen, Quy-Nhan Nguyen, and Minh-Tien Nguyen. 2026. ViLegalLM: Language Models for Vietnamese Legal Text. In Findings of the Association for Computational Linguistics: ACL 2026, pages 36136–36150, San Diego, California, United States. Association for Computational Linguistics.