中文
相关论文

相关论文: Denoising-based UNMT is more robust to word-order …

200 篇论文

Unsupervised neural machine translation (UNMT) has recently attracted great interest in the machine translation community. The main advantage of the UNMT lies in its easy collection of required large training text sentences while with only…

计算与语言 · 计算机科学 2020-12-04 Haipeng Sun , Rui Wang , Kehai Chen , Xugang Lu , Masao Utiyama , Eiichiro Sumita , Tiejun Zhao

Successful methods for unsupervised neural machine translation (UNMT) employ crosslingual pretraining via self-supervision, often in the form of a masked language modeling or a sequence generation task, which requires the model to align the…

计算与语言 · 计算机科学 2021-04-15 Alexandra Chronopoulou , Dario Stojanovski , Alexander Fraser

Unsupervised cross-lingual pretraining has achieved strong results in neural machine translation (NMT), by drastically reducing the need for large parallel data. Most approaches adapt masked-language modeling (MLM) to sequence-to-sequence…

计算与语言 · 计算机科学 2021-06-11 Christos Baziotis , Ivan Titov , Alexandra Birch , Barry Haddow

Despite pre-training's progress in many important NLP tasks, it remains to explore effective pre-training strategies for dense retrieval. In this paper, we propose RetroMAE, a new retrieval oriented pre-training paradigm based on Masked…

计算与语言 · 计算机科学 2022-10-18 Shitao Xiao , Zheng Liu , Yingxia Shao , Zhao Cao

Neural machine translation~(NMT) is ineffective for zero-resource languages. Recent works exploring the possibility of unsupervised neural machine translation (UNMT) with only monolingual data can achieve promising results. However, there…

计算与语言 · 计算机科学 2021-04-02 Mingxuan Wang , Hongxiao Bai , Hai Zhao , Lei Li

Masked auto-encoder pre-training has emerged as a prevalent technique for initializing and enhancing dense retrieval systems. It generally utilizes additional Transformer decoder blocks to provide sustainable supervision signals and…

信息检索 · 计算机科学 2024-04-23 Guangyuan Ma , Xing Wu , Zijia Lin , Songlin Hu

In this paper, we identify an interesting kind of error in the output of Unsupervised Neural Machine Translation (UNMT) systems like \textit{Undreamt}(footnote). We refer to this error type as \textit{Scrambled Translation problem}. We…

计算与语言 · 计算机科学 2021-06-18 Tamali Banerjee , Rudra Murthy , Pushpak Bhattacharyya

We present an ensemble-driven self-training framework for unsupervised neural machine translation (UNMT). Starting from a primary language pair, we train multiple UNMT models that share the same translation task but differ in an auxiliary…

计算与语言 · 计算机科学 2026-03-19 Ido Aharon , Jonathan Shaki , Sarit Kraus

For most language combinations, parallel data is either scarce or simply unavailable. To address this, unsupervised machine translation (UMT) exploits large amounts of monolingual data by using synthetic data generation techniques such as…

计算与语言 · 计算机科学 2021-07-20 Dana Ruiter , Dietrich Klakow , Josef van Genabith , Cristina España-Bonet

In self-supervised learning, it is challenging to reduce the gap between the enhancement performance on the estimated and target speech signals with existed pre-tasks. In this paper, we propose a multi-task pre-training method to improve…

声音 · 计算机科学 2022-01-02 Yi Li , Yang Sun , Syed Mohsen Naqvi

We propose Denoising Masked Autoencoder (Deno-MAE), a novel multimodal autoencoder framework for denoising modulation signals during pretraining. DenoMAE extends the concept of masked autoencoders by incorporating multiple input modalities,…

Text style transfer aims to change the style of sentences while preserving the semantic meanings. Due to the lack of parallel data, the Denoising Auto-Encoder (DAE) is widely used in this task to model distributions of different sentence…

计算与语言 · 计算机科学 2021-04-28 Jicheng Li , Yang Feng , Jiao Ou

Unsupervised neural machine translation (UNMT) has recently achieved remarkable results with only large monolingual corpora in each language. However, the uncertainty of associating target with source sentences makes UNMT theoretically an…

计算机视觉与模式识别 · 计算机科学 2019-05-28 Yuanhang Su , Kai Fan , Nguyen Bach , C. -C. Jay Kuo , Fei Huang

Pre-trained sequence-to-sequence models have significantly improved Neural Machine Translation (NMT). Different from prior works where pre-trained models usually adopt an unidirectional decoder, this paper demonstrates that pre-training a…

计算与语言 · 计算机科学 2022-06-03 Pengfei Li , Liangyou Li , Meng Zhang , Minghao Wu , Qun Liu

We present a simple and effective pretraining strategy {D}en{o}ising {T}raining DoT for neural machine translation. Specifically, we update the model parameters with source- and target-side denoising tasks at the early stage and then tune…

计算与语言 · 计算机科学 2022-01-21 Liang Ding , Keqin Peng , Dacheng Tao

Transfer learning approaches for Neural Machine Translation (NMT) train a NMT model on the assisting-target language pair (parent model) which is later fine-tuned for the source-target language pair of interest (child model), with the…

计算与语言 · 计算机科学 2019-04-11 Rudra Murthy , Anoop Kunchukuttan , Pushpak Bhattacharyya

Unsupervised learning of cross-lingual word embedding offers elegant matching of words across languages, but has fundamental limitations in translating sentences. In this paper, we propose simple yet effective methods to improve…

计算与语言 · 计算机科学 2019-01-08 Yunsu Kim , Jiahui Geng , Hermann Ney

Without real bilingual corpus available, unsupervised Neural Machine Translation (NMT) typically requires pseudo parallel data generated with the back-translation method for the model training. However, due to weak supervision, the pseudo…

计算与语言 · 计算机科学 2019-01-15 Shuo Ren , Zhirui Zhang , Shujie Liu , Ming Zhou , Shuai Ma

This paper demonstrates that multilingual denoising pre-training produces significant performance gains across a wide variety of machine translation (MT) tasks. We present mBART -- a sequence-to-sequence denoising auto-encoder pre-trained…

计算与语言 · 计算机科学 2020-01-24 Yinhan Liu , Jiatao Gu , Naman Goyal , Xian Li , Sergey Edunov , Marjan Ghazvininejad , Mike Lewis , Luke Zettlemoyer

This work presents a general unsupervised learning method to improve the accuracy of sequence to sequence (seq2seq) models. In our method, the weights of the encoder and decoder of a seq2seq model are initialized with the pretrained weights…

计算与语言 · 计算机科学 2018-02-23 Prajit Ramachandran , Peter J. Liu , Quoc V. Le
‹ 上一页 1 2 3 10 下一页 ›