Worlds Within Words: Translating Culture in Ancient Chinese Texts with Multi-Agent Coordination

2026-05-31

AI Summary

This study addresses the challenge of translating culture-loaded terms in Classical Chinese texts, where achieving both cultural explicitation and textual conciseness or readability is difficult. The authors propose MACAT, a novel framework that formulates this task as a selective explicitation problem and introduces, for the first time, a culturally aware multi-agent collaboration mechanism. By dynamically identifying culturally salient phrases and injecting concise explanatory knowledge, MACAT reconciles cultural fidelity with linguistic fluency. The framework incorporates quality-aware reranking and multi-round evaluation agents, leveraging large language models to optimize translation candidates. Experimental results on test sets from classical Chinese medical texts and the Analects demonstrate that MACAT significantly outperforms baseline models across multiple dimensions, including terminological accuracy, faithfulness, readability, and cultural preservation.

Abstract

Large language model (LLM)-based machine translation has advanced cross-cultural communication, yet it still struggles with culture-loaded words (CLWs) in ancient Chinese texts. The challenge extends beyond lexical alignment to deciding when and how culture-dependent knowledge should be explicated for readers lacking relevant background. Literal translation often preserves surface forms while missing underlying concepts, whereas over-explicitation harms conciseness and readability. To address this problem, we formulate CLW translation as a selective explicitation task and propose \textbf{MACAT}, a \textbf{M}ulti-\textbf{A}gent \textbf{C}ulture-\textbf{A}ware \textbf{T}ranslation framework that dynamically identifies culturally salient phrases and injects concise explanatory knowledge when necessary. MACAT further incorporates a quality-aware reranking module for candidate selection and a multi-round evaluation agent that assesses translations across terminological precision, readability, fidelity, cultural preservation, and cultural explicitation. Experiments on traditional Chinese medicine (TCM) classics and the \textit{Analects} show that, under a unified GPT-5.4 evaluation setting, MACAT consistently outperforms both the backbone model and general-purpose MT baselines on 100 TCM documents and a 20-chapter subset of the \textit{Analects}.

Problem

• culture-loaded words
• ancient Chinese texts
• machine translation
• cultural explicitation
• cross-cultural communication

Innovation

• multi-agent coordination
• culture-aware translation
• selective explicitation
• quality-aware reranking
• culturally loaded words