Translating Pro-Drop Languages With Reconstruction Models
Authors: Longyue Wang, Zhaopeng Tu, Shuming Shi, Tong Zhang, Yvette Graham, Qun Liu
AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on both Chinese English and Japanese English dialogue translation tasks show that the proposed approach significantly and consistently improves translation performance over a strong NMT baseline, which is directly built on the training data annotated with DPs. |
| Researcher Affiliation | Collaboration | Longyue Wang ADAPT Centre, Dublin City Univ. longyue.wang@adaptcentre.ie Zhaopeng Tu Tencent AI Lab zptu@tencent.com Shuming Shi Tencent AI Lab shumingshi@tencent.com Tong Zhang Tencent AI Lab bradymzhang@tencent.com Yvette Graham ADAPT Centre, Dublin City Univ. yvette.graham@adaptcentre.ie Qun Liu ADAPT Centre, Dublin City Univ. qun.liu@adaptcentre.ie |
| Pseudocode | No | No pseudocode or algorithm blocks are provided. |
| Open Source Code | No | Our released corpus is available at https://github.com/ longyuewangdcu/tvsub. |
| Open Datasets | Yes | 3. We release a large-scale bilingual dialogue corpus, which consists of 2.2M Chinese English sentence pairs.1 Our released corpus is available at https://github.com/ longyuewangdcu/tvsub. |
| Dataset Splits | Yes | We randomly select two complete television episodes as the tuning set, and another two episodes as the test set. |
| Hardware Specification | Yes | When running on a single GPU device Tesla K80, the training speed of the baseline model is 1.60K target words per second |
| Software Dependencies | No | We trained for 20 epochs using Adadelta (Zeiler 2012), and selected the model that yielded best performances on the tuning set. |
| Experiment Setup | Yes | The word-embedding dimension was 620 and the hidden layer size was 1,000. We trained for 20 epochs using Adadelta (Zeiler 2012), and selected the model that yielded best performances on the tuning set. The proposed model was implemented on top of the baseline model with the same settings where applicable. The hidden layer size in the reconstructor was 1,000. |