arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2312.00525cs.CLcs.AI
收藏

SurreyAI 2023 质量估计共享任务提交

SurreyAI 2023 Submission for the Quality Estimation Shared Task

University Of Surrey(萨里大学);Aston University(阿斯顿大学)

Archchana Sindhujan, Diptesh Kanojia, Constantin Orasan, Tharindu Ranasinghe

AI总结:

本文提出 SurreyAI 团队在 WMT23 句子级直接评估共享任务中的方法,基于 TransQuest 框架探索多种自编码器预训练语言模型,其中 MonoTQ-InfoXLM-large 在多数语言对上显著优于基线。

摘要:

质量估计(QE)系统在需要评估翻译质量但没有参考译文的情况下非常重要。本文描述了 SurreyAI 团队为解决 WMT23 句子级直接评估共享任务所采用的方法。所提出的方法建立在 TransQuest 框架之上,在 MonoTransQuest 架构中探索了多种自编码器预训练语言模型,并采用单模型和集成设置。所提出系统中使用的自编码器预训练语言模型包括 XLMV、InfoXLM-large 和 XLMR-large。评估采用 Spearman 和 Pearson 相关系数,考察机器预测质量分数与人工判断之间的关系,涵盖 5 个语言对(英语-古吉拉特语、英语-印地语、英语-马拉地语、英语-泰米尔语和英语-泰卢固语)。MonoTQ-InfoXLM-large 方法表现出稳健的策略,在大多数语言对上显著优于基线,超越了本研究中提出的所有其他单个模型。

英文摘要:

Quality Estimation (QE) systems are important in situations where it is necessary to assess the quality of translations, but there is no reference available. This paper describes the approach adopted by the SurreyAI team for addressing the Sentence-Level Direct Assessment shared task in WMT23. The proposed approach builds upon the TransQuest framework, exploring various autoencoder pre-trained language models within the MonoTransQuest architecture using single and ensemble settings. The autoencoder pre-trained language models employed in the proposed systems are XLMV, InfoXLM-large, and XLMR-large. The evaluation utilizes Spearman and Pearson correlation coefficients, assessing the relationship between machine-predicted quality scores and human judgments for 5 language pairs (English-Gujarati, English-Hindi, English-Marathi, English-Tamil and English-Telugu). The MonoTQ-InfoXLM-large approach emerges as a robust strategy, surpassing all other individual models proposed in this study by significantly improving over the baseline for the majority of the language pairs.

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。

↑