2019/07/19 by Pinjia He, Clara Meister, He, Pinjia +3 · 2 citations
Computer Science · #Artificial intelligence #Computer science #Dynamic and formal equivalence #Evaluation of machine translation #Example-based machine translation #Machine translation #Machine translation software usability #Natural language processing #Parsing #Phrase #Robustness (evolution) #Rule-based machine translation #Sentence #Software Engineering Research #Software Testing and Debugging Techniques #Syntax #Topic Modeling #Transfer-based machine translation #cs.CL #cs.SE
paper · pdf · doi:10.48550/arxiv.1907.08710
published in arXiv (Cornell University) (Cornell University) · Accepted at ICSE 2020
openalex publication_date 2019/07/19 · arxiv created 2020/07/14 · arxiv updated 2020/07/15 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
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.