# TransLaw: A Large-Scale Dataset and Multi-Agent Benchmark Simulating Professional Translation of Hong Kong Case Law

Source: [TransLaw: A Large-Scale Dataset and Multi-Agent Benchmark Simulating Professional Translation of Hong Kong Case Law](<https://www.sota2.com/research/paper/68fa79d01e209131cea7f832>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-20T05:00:05.764Z

Authors: Xi Xuan, Chunyu Kit

Published: 2025-07-01T15:39:26.000Z

## Abstract

Translating Hong Kong Court Judgments from English to Traditional Chinese is mandated by Articles 8-9 of the Basic Law, yet remains constrained by a shortage of parallel resources and rigorous demands on legal terminology, citation format, and judicial style. We introduce HKCFA Judgment 97-22, the first large-scale sentence-aligned parallel corpus for HK case law, comprising 344 professionally translated judgments (11,099 sentence pairs; 2.1M tokens) spanning 1997-2022. Building on this resource, we propose TransLaw, a multi-agent framework that decomposes translation into word-level expression, sentence-level translation, and multidimensional review, integrating a specialized Hong Kong legal glossary database, Retrieval-Augmented Generation, and iterative feedback, with four-dimensional expert review covering semantic alignment, terminology, citation, and style. Benchmarking 13 open-source and commercial LLMs, we demonstrate that TransLaw significantly outperforms single-agent baselines across all evaluated models, with convergence within 3 iterations. Human evaluation by 10 certified legal translators using our proposed Legal ACS metric confirms gains in legal-semantic accuracy, while showing that TransLaw still trails human experts in stylistic naturalness. The dataset and benchmark code are available at https://github.com/xuanxixi/TransLaw.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Legal Translation on HKCFA Judgment 97-22](<https://www.sota2.com/research/sota/legal-translation-on-hkcfa-judgment-97-22.md>): Legal Translation; HKCFA Judgment 97-22; xCOMET-XL: 85.12.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2507.00875>)

[PDF](<https://arxiv.org/pdf/2507.00875.pdf>)