論文閱讀：Fast Lexically Constrained Decoding with Dynamic Beam Allocation for Neural Machine Translation
內容簡介： 1.引言 詞法約束機器翻譯： 詞約束又叫做引導解碼，修改beam search 強迫輸出必須包含指定的詞或者短語。 雖然這個方法在理論上可行，但現有方法對於約束數量有線性（Hokamp和Liu，2017）或指數（Anderson等，2017）的計算複雜性。 這篇文章提出了一種用於詞約束解碼的算法，其約束數量爲O（1）的複雜度。並在sockeye上有實現。 詞法約束解碼實質上是對be
相關文章
- 1. 論文閱讀 Lexically Constrained Decoding for Sequence Generation Using Grid Beam Search
- 2. 論文閱讀：Dual Learning for Machine Translation
- 3. 論文閱讀：Diverse Beam Search--Decoding Diverse Solutions from Neural Sequence Models
- 4. 論文閱讀——Dual Learning for Machine Translation
- 5. 論文閱讀——Unsupervised Neural Machine Translation with Indirect Supervision
- 6. 論文閱讀——Modeling Future Cost for Neural Machine Translation
- 7. 論文閱讀——Robust Unsupervised Neural Machine Translation with Adversarial Training
- 8. 論文閱讀：Neural Machine Translation with Byte-Level Subwords
- 9. Neural machine translation by jointly learning to align and translate論文閱讀筆記
- 10. 【論文閱讀】Prior Knowledge Integration for Neural Machine Translation using Posterior Regularization
- 更多相關文章...
- • RSS 閱讀器 - RSS 教程
- • Docker Machine - Docker教程
- • JDK13 GA發佈：5大特性解讀
- • Scala 中文亂碼解決
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. 論文閱讀 Lexically Constrained Decoding for Sequence Generation Using Grid Beam Search
- 2. 論文閱讀：Dual Learning for Machine Translation
- 3. 論文閱讀：Diverse Beam Search--Decoding Diverse Solutions from Neural Sequence Models
- 4. 論文閱讀——Dual Learning for Machine Translation
- 5. 論文閱讀——Unsupervised Neural Machine Translation with Indirect Supervision
- 6. 論文閱讀——Modeling Future Cost for Neural Machine Translation
- 7. 論文閱讀——Robust Unsupervised Neural Machine Translation with Adversarial Training
- 8. 論文閱讀：Neural Machine Translation with Byte-Level Subwords
- 9. Neural machine translation by jointly learning to align and translate論文閱讀筆記
- 10. 【論文閱讀】Prior Knowledge Integration for Neural Machine Translation using Posterior Regularization