Neural Machine Translation (NMT) has made remarkable progress over the past
years. However, under-translation and over-translation remain two challenging
problems in state-of-the-art NMT systems. In this work, we conduct an in-depth
analysis on the underlying cause of under-translation in NMT, providing an
explanation from the perspective of decoding objective. To optimize the beam
search objective, the model tends to overlook words it is less confident about,
leading to the under-translation phenomenon. Correspondingly, the model's
confidence in predicting the End Of Sentence (EOS) diminishes when
under-translation occurs, serving as a mild penalty for under-translated
candidates. Building upon this analysis, we propose employing the confidence of
predicting EOS as a detector for under-translation, and strengthening the
confidence-based penalty to penalize candidates with a high risk of
under-translation. Experiments on both synthetic and real-world data show that
our method can accurately detect and rectify under-translated outputs, with
minor impact on other correct translations.

神经机器翻译 (NMT) 在过去几年取得了显著进展，但仍存在两个具有挑战性的问题，即欠翻译和过翻译。本文从解码目标的角度深入分析了 NMT 中欠翻译的根本原因，并提出了使用预测句子结束标记的置信度作为欠翻译检测器的方法，通过增强基于置信度的惩罚来纠正欠翻译的候选翻译，实验证明该方法可以准确地检测和纠正欠翻译的输出，对其他正确翻译的影响较小。