End-to-End Neural Word Alignment Outperforms GIZA++

Zenkel, Thomas; Wuebker, Joern; DeNero, John

doi:10.48550/arxiv.2004.14675

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GIZA++ [17], [18] is an implementation of IBM models. We used five iterations each for Model 1, the model, Model 3, and Model 4 to train GIZA++ by following the previous work of [19]. 3) AWE-SoME [20] is a neural word aligner based on multilingual BERT that can extract word alignments from contextualized word embeddings with and without fine-tuning on parallel data.…”

Section: ) Word Alignment Modelsmentioning

confidence: 99%

DbAPE: Denoising-Based APE System for Improving English-Myanmar NMT

2022

View full text Add to dashboard Cite

Automatic post-editing (APE) research aims to investigate methods for correcting systematic errors in machine translation (MT) results. Recent work has shown successful practices of APE for improving MT output quality; however, their effectiveness strongly relies on the availability of large-scale human-created APE triplets. The high production cost of human post-edited data has led to the absence of APE triplets for most language pairs, including English-Myanmar, which has become a limiting factor for the applicability of the APE task. This work investigates how to conduct the APE task on the English-Myanmar MT where human-edited APE triplets are unavailable. We build an APE system using only the monolingual and parallel MT corpora. The system takes the source sentence (src) and the MT output (mt) as inputs and produces the post-edited mt as output by operating the three processes together, including word alignment extraction, enriching mt using the extracted word alignment information, and denoising the enriched-version of mt. We conduct extensive experiments by applying our APE system as a post-processor to the raw output of the existing English-Myanmar MT systems and show that it significantly improves the quality of baseline translation results in terms of TER and BLEU scores. In addition, we perform word alignment experiments with four types of alignment methods and demonstrate that the proposed multilingual word aligner can achieve robust performance over previous state-of-the-art models.INDEX TERMS Automatic post editing, pre-trained embeddings, bilingual dictionary, word alignment, denoising.1 https://statmt.org/wmt21/ape-task.html

show abstract