重新审视用于跨语言分类的机器翻译
Revisiting Machine Translation for Cross-lingual Classification
AI总结：
本文重新审视跨语言分类中的机器翻译，表明通过使用更强的MT系统并缓解训练与推理文本不匹配，translate-test表现可大幅超越预期，最优方法高度依赖任务，挑战了多语言模型的主导地位。
摘要：
机器翻译（MT）已被广泛用于跨语言分类，方法包括将测试集翻译成英语并使用单语模型运行推理（translate-test），或将训练集翻译成目标语言并微调多语言模型（translate-train）。然而，该领域的大多数研究都集中在多语言模型上，而非MT组件。我们表明，通过使用更强大的MT系统，并缓解在原始文本上训练与在机器翻译文本上运行推理之间的不匹配问题，translate-test的表现可以大幅超越此前的预期。然而，最优方法高度依赖于任务，因为我们发现了影响不同任务和方法的跨语言迁移差距的各种来源。我们的工作对多语言模型在跨语言分类中的主导地位提出了质疑，并促使人们更加关注基于MT的基线方法。
英文摘要：
Machine Translation (MT) has been widely used for cross-lingual classification, either by translating the test set into English and running inference with a monolingual model (translate-test), or translating the training set into the target languages and finetuning a multilingual model (translate-train). However, most research in the area focuses on the multilingual models rather than the MT component. We show that, by using a stronger MT system and mitigating the mismatch between training on original text and running inference on machine translated text, translate-test can do substantially better than previously assumed. The optimal approach, however, is highly task dependent, as we identify various sources of cross-lingual transfer gap that affect different tasks and approaches differently. Our work calls into question the dominance of multilingual models for cross-lingual classification, and prompts to pay more attention to MT-based baselines.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
