paper-with-me

홈 › Papers

LawInstruct: A Resource for Studying Language Model Adaptation to the Legal Domain

2024-04-02 · Joel Niklaus, Lucia Zheng, Arya D. McCarthy, Christopher Hahn, Brian M. Rosen, Peter Henderson, Daniel E. Ho, Garrett Honke, Percy Liang, Christopher Manning

Instruction tuning is an important step in making language models useful for direct user interaction. However, the legal domain is underrepresented in typical instruction datasets (e.g., only 10 out of 1600+ tasks in Super-NaturalInstructions). To study whether instruction tuning on legal datasets is necessary for strong legal reasoning, we aggregate 58 annotated legal datasets and write instructions for each, creating LawInstruct. LawInstruct covers 17 global jurisdictions, 24 languages and a total of 12M examples across diverse tasks such as legal QA, summarization of court cases, and legal argument mining. We evaluate our models on LegalBench, measuring legal reasoning across five categories in 162 challenging and realistic legal tasks, and MMLU, to measure potential drops in general reasoning capabilities. We find that legal-specific instruction tuning on Flan-T5 - yielding FLawN-T5 - improves performance on LegalBench across all model sizes, with an aggregate increase of 15 points or 50% over Flan-T5 for the base size. No model size shows performance drops in MMLU. We publish LawInstruct as a resource for further study of instruction tuning in the legal domain.

📄 PDF Abstract BibTeX arXiv:2404.02127

Code (1)

joelniklaus/lawinstruct 공식 구현

Tasks

Argument MiningDecision MakingLanguage ModelingLanguage ModellingLegal ReasoningMMLU

Methods 이 논문이 사용한 방법론

BASE 설명 없음
Flan-T5 Flan-T5 is the instruction fine-tuned version of T5 or Text-to-Text Transfer Transformer Language Model.

Similar Papers 제목 키워드 기반

Mina: A Multilingual LLM-Powered Legal Assistant Agent for Bangladesh for Empowering Access to Justice

2025-11-04 · Azmine Toushik Wasi, Wahid Faisal, Mst Rafia Islam, Md Rizwan Parvez arxiv

Bangladesh's low-income population faces major barriers to affordable legal advice due to complex legal language, procedural opacity, and high costs. Existing AI legal assistants lack Bengali-language support and jurisdi…

Legal Reasoning

LegalCiteBench: Evaluating Citation Reliability in Legal Language Models

2026-05-11 · Sijia Chen, Hang Yin, Shunfan Zhou arxiv

Large language models (LLMs) are increasingly integrated into legal drafting and research workflows, where incorrect citations or fabricated precedents can cause serious professional harm. Existing legal benchmarks large…

Question Answering

JUÁ -- A Benchmark for Information Retrieval in Brazilian Legal Text Collections

2026-04-07 · Jayr Pereira, Leandro Fernandes, Erick de Brito, Roberto Lotufo 외 arxiv

Legal information retrieval in Portuguese remains difficult to evaluate systematically because available datasets differ widely in document type, query style, and relevance definition. We present JUÁ, a public benchmark …

Information RetrievalDomain Adaptation

Domain-Adaptation through Synthetic Data: Fine-Tuning Large Language Models for German Law

2026-01-20 · Ali Hamza Bashir, Muhammad Rehan Khalid, Kostadin Cvejoski, Jana Birr 외 arxiv

Large language models (LLMs) often struggle in specialized domains such as legal reasoning due to limited expert knowledge, resulting in factually incorrect outputs or hallucinations. This paper presents an effective met…

parameter-efficient fine-tuningSynthetic Data GenerationQuestion AnsweringLegal Reasoning

JuriBERT: A Masked-Language Model Adaptation for French Legal Text

2021-10-04 · EMNLP (NLLP) 2021 11 · Stella Douka, Hadi Abdine, Michalis Vazirgiannis, Rajaa El Hamdani 외

Language models have proven to be very useful when adapted to specific domains. Nonetheless, little research has been done on the adaptation of domain-specific BERT models in the French language. In this paper, we focus …

Language ModelingLanguage Modelling