Noisy Parallel Data Alignment
Not on the map. Why?
Abstract (arXiv)
An ongoing challenge in current natural language processing is how its major advancements tend to disproportionately favor resource-rich languages, leaving a significant number of under-resourced languages behind. Due to the lack of resources required to train and evaluate models, most modern language technologies are either nonexistent or unreliable to process endangered, local, and non-standardized languages. Optical character recognition (OCR) is often used to convert endangered language documents into machine-readable data. However, such OCR output is typically noisy, and most word alignment models are not built to work under such noisy conditions. In this work, we study the existing word-level alignment models under noisy settings and aim to make them more robust to noisy data. Our noise simulation and structural biasing method, tested on multiple language pairs, manages to reduce the alignment error rate on a state-of-the-art neural-based alignment model up to 59.6%.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 17 of its 46 references that are in the map. First posted on arXiv 23 Jan 2023.
Builds on
17 of 46 references.
References · 17
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2018Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyondcited by 434
- 2021Word Alignment by Fine-tuning Embeddings on Parallel Corporacited by 186
- 2020When Being Unseen from mBERT is just the Beginning: Handling New Languages With Multilingual Language Modelscited by 131
- 2019Jointly Learning to Align and Translate with Transformer Modelscited by 75
- 2020Accurate Word Alignment Induction from Neural Machine Translationcited by 32
- 2020OCR Post Correction for Endangered Language Textscited by 31
- 2020End-to-End Neural Word Alignment Outperforms GIZA++cited by 28
- 2020A Supervised Word Alignment Method based on Cross-Language Span Prediction using Multilingual BERTcited by 27
- 2019Low-Resource Corpus Filtering using Multilingual Sentence Embeddingscited by 25
- 2020Mask-Align: Self-Supervised Neural Word Alignmentcited by 20
- 2016Incorporating Structural Alignment Biases into an Attentional Neural Translation Modelcited by 16
5 more
- 2021Lexically Aware Semi-Supervised Learning for OCR Post-Correctioncited by 12
- 2018Part-of-Speech Tagging on an Endangered Language: a Parallel Griko-Italian Resourcecited by 10
- 2018Leveraging translations for speech transcription in low-resource settingscited by 5
- 2022OCR Improves Machine Translation for Low-Resource Languagescited by 5
- 2021MirrorAlign: A Super Lightweight Unsupervised Word Alignment Model via Cross-Lingual Contrastive Learningcited by 4
Built on by
Citing papers per year: 2023: 1.
1 paper.