论文标题
带有条件掩蔽的机器翻译推理策略
Inference Strategies for Machine Translation with Conditional Masking
论文作者
论文摘要
有条件的蒙版语言模型（CMLM）培训已被证明成功地用于非自动回归和半自动回调的序列生成任务，例如机器翻译。但是，鉴于训练有素的CMLM，尚不清楚最好的推理策略是什么。我们将掩盖的推论作为部分序列的条件概率分解，表明这不会损害性能，并研究了以这种观点激励的许多简单启发式方法。我们确定了与标准“面具预测”算法具有优势的阈值策略，并提供了其在机器翻译任务上的行为的分析。
Conditional masked language model (CMLM) training has proven successful for non-autoregressive and semi-autoregressive sequence generation tasks, such as machine translation. Given a trained CMLM, however, it is not clear what the best inference strategy is. We formulate masked inference as a factorization of conditional probabilities of partial sequences, show that this does not harm performance, and investigate a number of simple heuristics motivated by this perspective. We identify a thresholding strategy that has advantages over the standard "mask-predict" algorithm, and provide analyses of its behavior on machine translation tasks.
