已入深夜,您辛苦了!由于当前在线用户较少,发布求助请尽量完整地填写文献信息,科研通机器人24小时在线,伴您度过漫漫科研夜!祝你早点完成任务,早点休息,好梦!

Structure-Invariant Testing for Machine Translation

机器翻译 计算机科学 人工智能 判决 自然语言处理 机器翻译软件可用性 解析 基于迁移的机器翻译 短语 基于规则的机器翻译 机器翻译评价 基于实例的机器翻译 动态形式等价 稳健性(进化) 语法 基因 生物化学 化学
作者
Pinjia He,Clara Meister,Zhendong Su
出处
期刊:Cornell University - arXiv [Cornell University]
被引量:2
标识
DOI:10.48550/arxiv.1907.08710
摘要

In recent years, machine translation software has increasingly been integrated into our daily lives. People routinely use machine translation for various applications, such as describing symptoms to a foreign doctor and reading political news in a foreign language. However, the complexity and intractability of neural machine translation (NMT) models that power modern machine translation make the robustness of these systems difficult to even assess, much less guarantee. Machine translation systems can return inferior results that lead to misunderstanding, medical misdiagnoses, threats to personal safety, or political conflicts. Despite its apparent importance, validating the robustness of machine translation systems is very difficult and has, therefore, been much under-explored. To tackle this challenge, we introduce structure-invariant testing (SIT), a novel metamorphic testing approach for validating machine translation software. Our key insight is that the translation results of "similar" source sentences should typically exhibit similar sentence structures. Specifically, SIT (1) generates similar source sentences by substituting one word in a given sentence with semantically similar, syntactically equivalent words; (2) represents sentence structure by syntax parse trees (obtained via constituency or dependency parsing); (3) reports sentence pairs whose structures differ quantitatively by more than some threshold. To evaluate SIT, we use it to test Google Translate and Bing Microsoft Translator with 200 source sentences as input, which led to 64 and 70 buggy issues with 69.5\% and 70\% top-1 accuracy, respectively. The translation errors are diverse, including under-translation, over-translation, incorrect modification, word/phrase mistranslation, and unclear logic.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
A2ure完成签到 ,获得积分10
1秒前
Monica完成签到,获得积分10
1秒前
亚蛋求学完成签到,获得积分10
3秒前
yinh发布了新的文献求助10
4秒前
Soowu发布了新的文献求助10
5秒前
Jiayi发布了新的文献求助10
5秒前
快乐的寄容完成签到 ,获得积分10
5秒前
NexusExplorer应助维生素采纳,获得10
6秒前
7秒前
健壮凤凰完成签到,获得积分10
7秒前
7秒前
pwww发布了新的文献求助10
10秒前
Orange应助xiaopang采纳,获得10
11秒前
王wangdian发布了新的文献求助10
12秒前
12秒前
吴大王发布了新的文献求助10
13秒前
13秒前
威武的金毛完成签到 ,获得积分10
13秒前
小西贝完成签到 ,获得积分10
14秒前
16秒前
16秒前
虔虔发布了新的文献求助10
18秒前
从容清炎发布了新的文献求助10
18秒前
刘油果完成签到 ,获得积分10
18秒前
维生素发布了新的文献求助10
19秒前
jch完成签到 ,获得积分10
19秒前
BOBO发布了新的文献求助100
20秒前
azorworld6发布了新的文献求助10
22秒前
22秒前
科研通AI6.2应助azorworld6采纳,获得10
26秒前
科研通AI6.4应助虔虔采纳,获得20
27秒前
28秒前
29秒前
经久完成签到 ,获得积分10
31秒前
DrHHB应助XingHaiZHF采纳,获得50
31秒前
Jiayi发布了新的文献求助10
32秒前
azorworld6完成签到,获得积分10
33秒前
33秒前
34秒前
Lcx发布了新的文献求助10
40秒前
高分求助中
On lateral buckling of armouring wires in flexible pipes 10000
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
Navigating Normative Orders. Interdisciplinary Perspectives 800
Essentials of Carbohydrate Chemistry and Biochemistry, 4th Edition 700
1 Peter and Christ's Descent to the Dead in Its Early Christian Reception 700
Organizational Behavior 510
Management and the Arts 510
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 工程类 有机化学 化学工程 生物化学 计算机科学 内科学 物理 复合材料 催化作用 细胞生物学 无机化学 光电子学 物理化学 电极 基因
热门帖子
关注 科研通微信公众号,转发送积分 7744418
求助须知:如何正确求助?哪些是违规求助? 9292345
关于积分的说明 20211941
捐赠科研通 7323071
什么是DOI,文献DOI怎么找? 3307579
关于科研通互助平台的介绍 2459428
邀请新用户注册赠送积分活动 2318435