arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1809.07615cs.CL
收藏

多语言基础语言学习的经验教训

Lessons learned in multilingual grounded language learning

Tilburg University(蒂尔堡大学);University of Copenhagen(哥本哈根大学);Microsoft Research Montreal(微软研究院蒙特利尔);University of Edinburgh(爱丁堡大学)

Ákos Kádár, Desmond Elliott, Marc-Alexandre Côté, Grzegorz Chrupała, Afra Alishahi

摘要:

最近的工作已经展示了如何通过利用多种语言的图像描述来学习更好的视觉语义嵌入。在这里,我们详细调查哪些条件影响这种类型的接地语言学习模型的性能。我们表明,多语言培训比双语培训更好,并且低资源语言从高资源语言的培训中受益。我们证明,多语言模型可以在翻译或可比句子对上同样良好地训练,并且用多种语言注释同一组图像可以通过额外的标题-标题排名目标进行进一步改进。

英文摘要:

Recent work has shown how to learn better visual-semantic embeddings by leveraging image descriptions in more than one language. Here, we investigate in detail which conditions affect the performance of this type of grounded language learning model. We show that multilingual training improves over bilingual training, and that low-resource languages benefit from training with higher-resource languages. We demonstrate that a multilingual model can be trained equally well on either translations or comparable sentence pairs, and that annotating the same set of images in multiple language enables further improvements via an additional caption-caption ranking objective.

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。

↑