arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.03075cs.CL
收藏

面向多模态机器翻译的具体视觉标记检测

Detecting Concrete Visual Tokens for Multimodal Machine Translation

PAR Government Systems Corporation(PAR政府系统公司);Air Force Research Laboratory(空军研究实验室)

Braeden Bowen, Vipin Vijayan, Scott Grigsby, Timothy Anderson, Jeremy Gwinnup

AI总结:

针对多模态机器翻译的视觉接地与掩码挑战,提出多种具体视觉标记检测与选择方法,基于GRAM架构在合成多模态数据集上训练,实现了性能与视觉语境利用率的提升。

摘要:

多模态机器翻译(MMT)系统中视觉接地与掩码的挑战推动了多种面向掩码的视觉接地文本标记检测与选择方法的发展。本文提出了从源语句中检测视觉与语境相关(具体)标记的新方法,包括基于自然语言处理(NLP)的检测、基于目标检测的检测,以及一种联合检测-验证技术。本文还提出了已检测标记的新选择方法,包括最短n个标记、最长n个标记,以及所有检测到的具体标记。本文采用GRAM MMT架构,在由源图像与掩码语句合成整理的多模态数据集上训练模型,结果表明相较于基线模型,其在翻译任务中性能有所提升,且对视觉语境的利用更充分。

英文摘要:

The challenge of visual grounding and masking in multimodal machine translation (MMT) systems has encouraged varying approaches to the detection and selection of visually-grounded text tokens for masking. We introduce new methods for detection of visually and contextually relevant (concrete) tokens from source sentences, including detection with natural language processing (NLP), detection with object detection, and a joint detection-verification technique. We also introduce new methods for selection of detected tokens, including shortest $n$ tokens, longest $n$ tokens, and all detected concrete tokens. We utilize the GRAM MMT architecture to train models against synthetically collated multimodal datasets of source images with masked sentences, showing performance improvements and improved usage of visual context during translation tasks over the baseline model.

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。

↑