{
 "id": "ntphuc149/ViLegalTF",
 "owner": "ntphuc149",
 "name": "ViLegalTF",
 "downloads": 11,
 "likes": 2,
 "last_modified": "2026-06-29T05:20:42.000Z",
 "stated_licence": "apache-2.0",
 "stated_from": "tag",
 "licence_name": null,
 "licence_link": null,
 "file_licence": null,
 "licence_files": [],
 "sources": [],
 "gated": "auto",
 "tasks": [
  "question-answering"
 ],
 "citation": "@inproceedings{nguyen-etal-2026-vilegallm,\n    title = \"{V}i{L}egal{LM}: Language Models for {V}ietnamese Legal Text\",\n    author = \"Nguyen, Truong-Phuc  and\n      Nguyen, Quy-Nhan  and\n      Nguyen, Minh-Tien\",\n    editor = \"Liakata, Maria  and\n      Moreira, Viviane P.  and\n      Zhang, Jiajun  and\n      Jurgens, David\",\n    booktitle = \"Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026\",\n    month = jul,\n    year = \"2026\",\n    address = \"San Diego, California, United States\",\n    publisher = \"Association for Computational Linguistics\",\n    url = \"https://aclanthology.org/2026.findings-acl.1801/\",\n    pages = \"36136--36150\",\n    ISBN = \"979-8-89176-395-1\",\n    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 continu",
 "signal": "permissive",
 "reason": "The declared licence (apache-2.0) is classified as permissive.",
 "revision": "9aa685a288689d89c2fde838b782524964605a41",
 "stated_evidence": "https://huggingface.co/datasets/ntphuc149/ViLegalTF/blob/9aa685a288689d89c2fde838b782524964605a41/README.md",
 "file_evidence": "https://huggingface.co/datasets/ntphuc149/ViLegalTF/tree/9aa685a288689d89c2fde838b782524964605a41",
 "card_nc": false
}