Papers by Brian Thompson
ParaCrawl: Web-Scale Acquisition of Parallel Corpora (2020.acl-main)
Copied to clipboard
Marta Bañón, Pinzhen Chen, Barry Haddow, Kenneth Heafield, Hieu Hoang, Miquel Esplà-Gomis, Mikel L. Forcada, Amir Kamran, Faheem Kirefu, Philipp Koehn, Sergio Ortiz Rojas, Leopoldo Pla Sempere, Gema Ramírez-Sánchez, Elsa Sarrías, Marek Strelec, Brian Thompson, William Waites, Dion Wiggins, Jaume Zaragoza
| Challenge: | We describe methods to create the largest publicly available parallel corpora by crawling the web . parallel corpus is essential for building highquality machine translation systems . |
| Approach: | They describe methods to create largest publicly available parallel corpora by crawling web sites . they empirically compare alternative methods and publish benchmark data sets . |
| Outcome: | The proposed methods improve state-of-the-art results on common benchmarks, the authors show . the pipeline has been tested on Russian, Sinhala, Nepali, Tagalog, Swahili, and Somali . |
A Shocking Amount of the Web is Machine Translated: Insights from Multi-Way Parallelism (2024.findings-acl)
Copied to clipboard
| Challenge: | Multi-way parallel, machine generated content dominates the translations in lower resource languages . a limited investigation suggests this selection bias is the result of low quality content generated in English and translated into many lower resource language via MT. |
| Approach: | They show that multi-way parallel, machine generated content dominates translations in many languages . they also find evidence of a selection bias in the type of content which is translated into many languages. |
| Outcome: | The results suggest that the low quality of multi-way translations on the web was likely created using machine translation. |
End-to-End Single-Channel Speaker-Turn Aware Conversational Speech Translation (2023.emnlp-main)
Copied to clipboard
Juan Pablo Zuluaga-Gomez, Zhaocheng Huang, Xing Niu, Rohit Paturi, Sundararajan Srinivasan, Prashant Mathur, Brian Thompson, Marcello Federico
| Challenge: | Conventional speech-to-text translation systems are trained on single-speaker utterances, but they may not be applicable to real-life scenarios where the audio contains conversations by multiple speakers. |
| Approach: | They propose a speaker-turn-aware conversational speech translation model that integrates automatic speech recognition, speech translation and speaker turn detection using special tokens in a serialized labeling format. |
| Outcome: | The proposed model outperforms the reference systems on the multi-speaker condition while attaining comparable performance on the single-speakspeaker conditions. |
HABLex: Human Annotated Bilingual Lexicons for Experiments in Machine Translation (D19-1)
Copied to clipboard
| Challenge: | Existing methods to incorporate bilingual lexicons into statistical machine translation are unclear how to do so in the neural framework. |
| Approach: | They present a dataset to test methods for bilingual lexicon integration into neural machine translation using human generated alignments of words and phrases in three language pairs. |
| Outcome: | The proposed method improves on baselines and improves training to address overfitting. |
Automatic Machine Translation Evaluation in Many Languages via Zero-Shot Paraphrasing (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing metrics for machine translation evaluation are causing the correlation between human judgments and automatic metrics to break down. |
| Approach: | They propose to train a multilingual NMT system to score machine translation output with a sequence-to-sequence paraphraser, conditioned on a human reference. |
| Outcome: | The proposed model outperforms or statistically ties with all prior metrics on the WMT 2019 segment-level shared metrics task in all languages (excluding Gujarati where the model had no training data). |
Overcoming Catastrophic Forgetting During Domain Adaptation of Neural Machine Translation (N19-1)
Copied to clipboard
| Challenge: | Neural Machine Translation (NMT) performs poorly without large training corpora. |
| Approach: | They propose a machine learning method that retains the majority of general-domain performance lost in continued training without degrading in-domain. |
| Outcome: | The proposed method retains the majority of general-domain performance lost in continued training without degrading in-domain performances. |
Benchmarking Neural and Statistical Machine Translation on Low-Resource African Languages (2020.lrec-1)
Copied to clipboard
| Challenge: | a recent study has focused on languages where large amounts of resources are available. |
| Approach: | They benchmark state of the art statistical and neural machine translation systems on Somali and Swahili languages . they find that statistical machine translation and neural translation can perform similarly in low-resource scenarios . |
| Outcome: | The results show that statistical machine translation and neural machine translation perform similarly in low-resource scenarios. |
Simulated multiple reference training improves low-resource machine translation (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing valid translations for a given sentence are limited by a single reference translation, causing data sparsity in low-resource settings. |
| Approach: | They propose a method that approximates the full space of possible translations by sampling a paraphrase of the reference sentence from a MT model and training it to predict the paraphraser’s distribution over possible tokens. |
| Outcome: | The proposed method improves in low-resource settings and is complementary to back-translation. |
Improving Retrieval Augmented Neural Machine Translation by Controlling Source and Fuzzy-Match Interactions (2023.findings-eacl)
Copied to clipboard
| Challenge: | a general-domain model has access to customer or domain specific parallel data at inference time, but not during training. |
| Approach: | They propose a zero-shot adaptation approach where a general-domain model has access to customer or domain specific parallel data at inference time, but not during training. |
| Outcome: | The proposed architecture outperforms existing architectures in two language pairs . it consistently improves BLEU across language pair, domain, and number k of fuzzy matches . |
Exploiting Sentence Order in Document Alignment (2020.emnlp-main)
Copied to clipboard
| Challenge: | a document alignment method that exploits sentence order information is beneficial even when the end goal is sentence-level bitext. |
| Approach: | They propose a document alignment method that incorporates sentence order information in both candidate generation and candidate re-scoring. |
| Outcome: | The proposed method outperforms the most recent document alignment method on Sinhala–English documents. |
Dubbing in Practice: A Large Scale Study of Human Localization With Insights for Automatic Dubbing (2023.tacl-1)
Copied to clipboard
| Challenge: | a large-scale study of human dubbing in practice is lacking in qualitative literature on human dubs . authors argue for vocal naturalness and translation quality over isometric constraints . a data-driven examination of the way humans perform this task is needed . |
| Approach: | They analyze 319.57 hours of video from 54 professionally produced titles . they argue for vocal naturalness and translation quality over isometric constraints . authors say they need to preserve speech characteristics and transfer of semantic properties . |
| Outcome: | The study challenges assumptions in qualitative and machine-learning literature on dubbing . it also finds that source-side audio influences human dubbing through other channels . |
Fine-Tuned Machine Translation Metrics Struggle in Unseen Domains (2024.acl-short)
Copied to clipboard
| Challenge: | a new dataset examines whether fine-tuned metrics are robust to domain shifts between training and inference. |
| Approach: | They use an annotated multidimensional quality metrics dataset to examine whether they are robust to domain shifts between training and inference. |
| Outcome: | The proposed metrics exhibit a substantial performance drop in the unseen domain scenario compared to metrics that rely on the surface form and pre-trained metrics that are not fine-tuned on MT quality judgments. |
Vecalign: Improved Sentence Alignment in Linear Time and Space (D19-1)
Copied to clipboard
| Challenge: | Sentence-aligned bitext is used to train nearly all machine translation systems. |
| Approach: | They propose a bilingual sentence alignment method which is linear in time and space with respect to the number of sentences being aligned. |
| Outcome: | The proposed method outperforms the existing method by 5 F1 points on a German–French test set and improves downstream MT quality by 1.7 and 1.6 BLEU in Sinhala-English and Nepali-English, respectively. |