构建阿拉伯语-俄语科学翻译语料库，助力跨语言科研协作。
Bridging Scientific Heritage: An Arabic--Russian Parallel Corpus and LLM Benchmark for Sustainable Knowledge Transfer
- 整合2.7万句对科学文本，构建跨语言平行语料库。
- 70亿参数模型经微调后，翻译指标提升4.36点以上。
- 支持可持续发展研究合作，适合多语言AI研究者使用。
俄语和阿拉伯语是重要的科学交流语言。语言障碍阻碍了这两个学术群体间的研究成果共享，影响国际协作及可持续性研究进展。本文提出一个阿拉伯语-俄语科学翻译基准，包含约2.7万句对的混合平行语料，涵盖科学摘要与通用领域文本（宗教、新闻、对话）。我们使用LoRA对三个多语言模型（mT5-base, NLLB-200-distilled-1.3B, Qwen2.5-7B-Instruct）进行微调，秩分别为8、16、32、64。其中，70亿参数的Qwen2.5-7B模型采用QLoRA（秩8）时，取得BLEU 23.15、chrF 43.89、BERTScore 0.906、COMET 0.758，较零样本基线分别提升+4.36（BLEU）和+0.051（COMET）。三例少样本提示未提升性能，表明需专用领域微调。我们公开模型、语料与评估代码。该工作降低科学文献语言壁垒，促进阿拉伯语与俄语科研人员的知识互通，助力可持续伙伴关系（联合国可持续发展目标17）与创新基础设施建设（目标9），契合技术驱动可持续发展的会议主题。
原文摘要 · Abstract (English)
Russian and Arabic are among the major languages of scientific communication. Language barriers impede the exchange of research results between these communities, which affects international collaboration and the progress of sustainability-related research. We present a benchmark for Arabic--Russian scientific translation. The benchmark includes a hybrid parallel corpus of about 27,000 sentence pairs, compiled from scientific abstracts and general-domain texts (religion, news, conversations). We fine-tune three multilingual language models -- mT5-base (580M parameters), NLLB-200-distilled-1.3B (1.3B), and Qwen2.5-7B-Instruct (7B) -- using LoRA with ranks 8, 16, 32, and 64. The Qwen2.5-7B model with QLoRA (rank 8) yields BLEU 23.15, chrF 43.89, BERTScore 0.906, and COMET 0.758. These are +4.36 BLEU and +0.051 COMET above the zero-shot baseline. Few-shot prompting with three examples does not improve performance, indicating that domain-specific fine-tuning is required. We release the models, the corpus, and the evaluation code. By lowering the language barrier for scientific texts, the work enables knowledge exchange between Arabic-speaking and Russian-speaking researchers. It contributes to sustainable partnerships (UN SDG 17) and innovation infrastructure (SDG 9), aligning with the conference's focus on technology-driven sustainable development.
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。