AI Paper Atlas

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

Translating Pro-Drop Languages With Reconstruction Models

Longyue Wang, Zhaopeng Tu, Shuming Shi, Tong Zhang, Yvette Graham, Qun Liu

AAAI 2018 · Label-efficient and robust learning › Recognition and long-tail · approach: Multilingual language representation pretraining · 46 citations (OpenAlex)

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 encoder-side and decoder-side reconstructors of DP-labelled sourceNMT with DP-labelled training data · Chinese–English dialogueBLEU: 32.67 -> 35.08Improved
Add encoder-side reconstructionNMT with DP-labelled training data · Chinese–English dialogueBLEU: 32.67 -> 33.67Improved
Add decoder-side reconstructionNMT with DP-labelled training data · Chinese–English dialogueBLEU: 32.67 -> 33.48Improved
Train on explicitly DP-labelled source sentencesNMT baseline · Chinese–English dialogueBLEU: 31.80 -> 32.67Improved
Add encoder-side and decoder-side DP reconstructionNMT with DP-labelled training data · Japanese–EnglishBLEU: 20.55 -> 21.84Improved
Do not use reconstruction during testing; retain it during training (ablation)DP reconstruction with encoder and decoder reconstructors · Chinese–English dialogueBLEU: 32.67 -> 34.02Improved
Reconstruct the original source instead of the DP-labelled source (ablation)NMT reconstruction with encoder and decoder reconstructors · Chinese–English dialogueBLEU: 35.08 -> 33.25Hurt
Use manual rather than automatic DP labels at test time (ablation)DP reconstruction with encoder and decoder reconstructors · Chinese–English dialogueBLEU: 35.08 -> 38.38Improved
Increase the hidden layer size to 2100 without adding DP reconstruction (ablation)Baseline (+DPs) · Chinese–English dialogueBLEU: 32.67 -> 32.00No clear effect
Use a three-layer encoder and decoder instead of reconstruction (ablation)Baseline (+DPs) · Chinese–English dialogueBLEU: 33.36 vs. 35.08Hurt

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