paper-with-me

홈 › Papers

Modeling Dual Read/Write Paths for Simultaneous Machine Translation

2022-03-17 · ACL 2022 5 · Shaolei Zhang, Yang Feng

Simultaneous machine translation (SiMT) outputs translation while reading source sentence and hence requires a policy to decide whether to wait for the next source word (READ) or generate a target word (WRITE), the actions of which form a read/write path. Although the read/write path is essential to SiMT performance, no direct supervision is given to the path in the existing methods. In this paper, we propose a method of dual-path SiMT which introduces duality constraints to direct the read/write path. According to duality constraints, the read/write path in source-to-target and target-to-source SiMT models can be mapped to each other. As a result, the two SiMT models can be optimized jointly by forcing their read/write paths to satisfy the mapping. Experiments on En-Vi and De-En tasks show that our method can outperform strong baselines under all latency.

📄 PDF Abstract BibTeX arXiv:2203.09163

Code (1)

ictnlp/dual-paths 공식 구현 pytorch

Tasks

de-enMachine TranslationSentenceTranslation

Similar Papers 제목 키워드 기반

Self-Modifying State Modeling for Simultaneous Machine Translation

2024-06-04 · Donglei Yu, Xiaomian Kang, Yuchen Liu, Yu Zhou 외

Simultaneous Machine Translation (SiMT) generates target outputs while receiving stream source inputs and requires a read/write policy to decide whether to wait for the next source token or generate a new target token, w…

Machine TranslationTranslation

Eyettention: An Attention-based Dual-Sequence Model for Predicting Human Scanpaths during Reading

2023-04-21 · Shuwen Deng, David R. Reich, Paul Prasse, Patrick Haller 외

Eye movements during reading offer insights into both the reader's cognitive processes and the characteristics of the text that is being read. Hence, the analysis of scanpaths in reading have attracted increasing attenti…

Cross Attention Augmented Transducer Networks for Simultaneous Translation

2021-11-01 · EMNLP 2021 11 · Dan Liu, Mengge Du, Xiaoxi Li, Ya Li 외

This paper proposes a novel architecture, Cross Attention Augmented Transducer (CAAT), for simultaneous translation. The framework aims to jointly optimize the policy and translation models. To effectively consider all p…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Decoderspeech-recognition+3

Residual Networks Behave Like Ensembles of Relatively Shallow Networks

2016-05-20 · NeurIPS 2016 12 · Andreas Veit, Michael Wilber, Serge Belongie

In this work we propose a novel interpretation of residual networks showing that they can be seen as a collection of many paths of differing length. Moreover, residual networks seem to enable very deep networks by levera…

Non-Monotonic Attention-based Read/Write Policy Learning for Simultaneous Translation

2025-03-28 · Zeeshan Ahmed, Frank Seide, Zhe Liu, Rastislav Rabatin 외

Simultaneous or streaming machine translation generates translation while reading the input stream. These systems face a quality/latency trade-off, aiming to achieve high translation quality similar to non-streaming mode…

Binary ClassificationMachine TranslationTranslation