تقرير
Why Neural Machine Translation Prefers Empty Outputs
| العنوان: | Why Neural Machine Translation Prefers Empty Outputs |
|---|---|
| المؤلفون: | Shi, Xing, Xiao, Yijun, Knight, Kevin |
| سنة النشر: | 2020 |
| المجموعة: | Computer Science |
| مصطلحات موضوعية: | Computer Science - Computation and Language |
| الوصف: | We investigate why neural machine translation (NMT) systems assign high probability to empty translations. We find two explanations. First, label smoothing makes correct-length translations less confident, making it easier for the empty translation to finally outscore them. Second, NMT systems use the same, high-frequency EoS word to end all target sentences, regardless of length. This creates an implicit smoothing that increases zero-length translations. Using different EoS types in target sentences of different lengths exposes and eliminates this implicit smoothing. Comment: 6 pages |
| نوع الوثيقة: | Working Paper |
| URL الوصول: | http://arxiv.org/abs/2012.13454 |
| رقم الأكسشن: | edsarx.2012.13454 |
| قاعدة البيانات: | arXiv |
| الوصف غير متاح. |