研究长文档翻译性能下降问题，发现越长越差且开头句子译得更好。
Investigating Length Issues in Document-level Machine Translation
- 设计新方法精确测量长度对翻译的影响
- 输入越长翻译质量越差，开头句子表现更优
- 调整长度分布和位置嵌入效果有限，仍不如句级翻译
Transformer 架构在处理和生成超长文本方面日益有效，为文档级机器翻译（MT）开辟了新前景。本文挑战了 MT 系统处理长达数千词文本的能力。我们设计并实现了一种新方法，精准衡量长度增加对翻译输出的影响。在两种代表性架构上的实验明确显示：(a) 翻译性能随输入文本长度增加而下降；(b) 句子在文档中的位置具有影响，靠前的句子翻译质量更高。我们进一步发现，仅调整文档长度分布和位置嵌入分布，对这些问题的缓解作用微乎其微。结果表明，尽管文档级 MT 在计算上可行，但其性能尚未达到句级翻译水平。
原文摘要 · Abstract (English)
Transformer architectures are increasingly effective at processing and generating very long chunks of texts, opening new perspectives for document-level machine translation (MT). In this work, we challenge the ability of MT systems to handle texts comprising up to several thousands of tokens. We design and implement a new approach designed to precisely measure the effect of length increments on MT outputs. Our experiments with two representative architectures unambiguously show that (a)~translation performance decreases with the length of the input text; (b)~the position of sentences within the document matters, and translation quality is higher for sentences occurring earlier in a document. We further show that manipulating the distribution of document lengths and of positional embeddings only marginally mitigates such problems. Our results suggest that even though document-level MT is computationally feasible, it does not yet match the performance of sentence-based MT.
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。