AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

“Bilingual Expert” Can Find Translation Errors

Kai Fan, Jiayi Wang, Bo Li, Fengming Zhou, Boxing Chen, Luo Si

AAAI 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Multilingual language representation pretraining

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add model-derived and mis-matching features from a pretrained bilingual expertBi-LSTM sentence QE · WMT 2017 en-de and de-enPearson's: 0.6599 -> 0.6837; 0.6985 -> 0.7099Improved
Remove mis-matching features (ablation)Ours Single sentence QE · WMT 2017 en-de testPearson's: 0.6837 -> 0.6763Hurt
Add mis-matching features (ablation)Ours Single sentence QE · WMT 2017 de-en testPearson's: 0.7063 -> 0.7099Improved
Remove model-derived features (ablation)Ours Single sentence QE · WMT 2017 en-de testPearson's: 0.6837 -> 0.6408Hurt
Add model-derived features (ablation)Ours Single sentence QE · WMT 2017 de-en testPearson's: 0.6726 -> 0.7099Improved
Use the proposed ensemble QE systemSentence-level QE · WMT 2018 en-de and de-enPearson's: 0.7000 -> 0.7308; 0.7667 -> 0.7631Mixed
Use the proposed system instead of the strongest listed baselineWord-level QE · WMT 2017 en-de and de-enF1-Multi: 0.568 -> 0.5826; 0.535 -> 0.5507Improved
Use BPE rather than word tokenization (ablation)Quality estimation · Most evaluated sentence- and word-level tasksQE performanceImproved
Replace Bi-LSTM with a self-attention module (ablation)Quality estimation predictor · QE experimentsQE performanceHurt
Add a CRF layer to the Bi-LSTM (ablation)Word-level QE predictor · QE experimentsQE performanceNo clear effect
Use WMT parallel data instead of CWMT-only parallel data (ablation)Single-model bilingual expert for CWMT QE · CWMT 2018 Ch->En and En->ChPearson's: 0.564 -> 0.612; 0.588 -> 0.620Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy