English(EN) Back-Translation Still Matters: Leveraging Monolingual Data for Low-Resource Machine Translation with LLMs

LLM 的反向翻译可提升低资源机器翻译

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-15 04:00

一篇新的研究论文探讨了在有大型语言模型 (LLM) 可用的情况下，反向翻译对于低资源机器翻译的有效性。该研究以奥里亚语和斯瓦希里语等语言为重点，发现反向翻译的性能明显优于持续预训练，甚至优于拥有更多并行数据的模型。研究还强调了单语文本的质量和来源对于获得最佳结果的重要性。 AI

影响增强了服务不足语言的机器翻译能力，可能改善全球信息获取和沟通。

排序理由研究论文，详细介绍了该技术（反向翻译）在改进 LLM 在特定任务（低资源机器翻译）上的性能方面的新应用。[lever_c_demoted from research: ic=1 ai=1.0]

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 16 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 研究论文，详细介绍了该技术（反向翻译）在改进 LLM 在特定任务（低资源机器翻译）上的性能方面的新应用。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

arXiv cs.AI TIER_1 English(EN) · Divyanshu Bathla, Parag Singla, Jitin Singla · 2026-10-15 04:00

回译仍有价值：利用单语数据通过LLM提升低资源机器翻译

arXiv:2610.18818v2 Announce Type: cross Abstract: Pre-trained Large Language Model (LLMs) are effective on many tasks in zero-shot and few-shot settings. For machine translation of low-resource languages, the obvious options are few-shot prompting and fine-tuning on a small paral…