“Bilingual Expert” Can Find Translation Errors

Authors: Kai Fan, Jiayi Wang, Bo Li, Fengming Zhou, Boxing Chen, Luo Si6367-6374

AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The experimental results show that our approach achieves the state-of-the-art performance in most public available datasets of WMT 2017/2018 QE task.
Researcher Affiliation Industry Kai Fan, Jiayi Wang, Bo Li, Fengming Zhou, Boxing Chen, Luo Si Alibaba Group Inc. k.fan,joanne.wjy,shiji.lb,zfm104435,boxing.cbx,luo.si@alibaba-inc.com
Pseudocode Yes Algorithm 1 Translation Quality Estimation with Bi Transformer and Bi-LSTM
Open Source Code No The paper does not provide a link to its source code or explicitly state that the code for their methodology is open-source or publicly available.
Open Datasets Yes The data resources that we used for training the neural Bilingual Expert model are mainly from WMT1: (i) parallel corpora released for the WMT17/18 News Machine Translation Task, (ii) UFAL Medical Corpus and Khresmoi development data released for the WMT17/18 Biomedical Translation Task, (iii) src-pe pairs for the WMT17/18 QE Task. 1http://www.statmt.org/wmt18/
Dataset Splits Yes We evaluate our algorithm on the testing data of WMT 2017/2018, and development data of CWMT 2018. For fair comparison, we tuned all the hyper-parameters of our model on the development data, and reported the corresponding results for the testing data.
Hardware Specification Yes The bilingual expert model is trained on 8 Nvidia P-100 GPUs for about 3 days until convergence. For translation QE model, we use only one layer Bi-LSTM, and it is trained on a single GPU.
Software Dependencies No The paper mentions software like 'scikit-learn' and 'CRFSuite toolkit' but does not specify their version numbers or other required software dependencies with versions.
Experiment Setup Yes The number of layers in the bidirectional transformer for each module is 2, and the number of hidden units for feedforward sub-layer is 512. We use the 8-head self-attention in practice, since the single one is just a weighted average of previous layers. For translation QE model, we use only one layer Bi-LSTM...