通过自我反思提升低资源翻译质量，无需微调。
Reflective Translation: Improving Low-Resource Machine Translation via Structured Self-Reflection
- 先生成译文，再结构化自省，最后优化输出。
- 平均提升0.22 BLEU和0.18 COMET，效果显著。
- 适合低资源语言研究者，可直接部署于大模型。
如祖鲁语和科萨语等低资源语言在机器翻译中因平行数据和语言资源有限而面临持续挑战。近期大型语言模型的发展表明，自我反思——即模型对自身输出进行批判与修正——可提升推理质量和事实一致性。本文提出反射式翻译（Reflective Translation），一种基于提示的框架：模型先生成初始译文，再生成结构化自省，最后利用该反思生成优化译文。在英语-祖鲁语和英语-科萨语翻译任务上，使用OPUS-100和NTREX-African数据集，评估了多种提示策略与置信度阈值。结果表明，第一轮与第二轮译文在BLEU和COMET得分上均有稳定提升，平均分别提高+0.22和+0.18。配对非参数检验确认提升具有统计显著性。该方法不依赖具体模型，无需微调，并生成可用于后续监督或分析工作的反思增强数据集。研究证明，结构化自我反思是改善低资源翻译质量的有效且实用机制。
原文摘要 · Abstract (English)
Low-resource languages such as isiZulu and isiXhosa face persistent challenges in machine translation due to limited parallel data and linguistic resources. Recent advances in large language models suggest that self-reflection, prompting a model to critique and revise its own outputs, can improve reasoning quality and factual consistency. Building on this idea, this paper introduces Reflective Translation, a prompt-based framework in which a model generates an initial translation, produces a structured self-critique, and then uses this reflection to generate a refined translation. The approach is evaluated on English-isiZulu and English-isiXhosa translation using OPUS-100 and NTREX-African, across multiple prompting strategies and confidence thresholds. Results show consistent improvements in both BLEU and COMET scores between first- and second-pass translations, with average gains of up to +0.22 BLEU and +0.18 COMET. Statistical significance testing using paired nonparametric tests confirms that these improvements are robust. The proposed method is model-agnostic, requires no fine-tuning, and introduces a reflection-augmented dataset that can support future supervised or analysis-driven work. These findings demonstrate that structured self-reflection is a practical and effective mechanism for improving translation quality in low-resource settings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。