paper-with-me

홈 › Papers

Can Latent Alignments Improve Autoregressive Machine Translation?

2021-04-19 · NAACL 2021 4 · Adi Haviv, Lior Vassertail, Omer Levy

Latent alignment objectives such as CTC and AXE significantly improve non-autoregressive machine translation models. Can they improve autoregressive models as well? We explore the possibility of training autoregressive machine translation models with latent alignment objectives, and observe that, in practice, this approach results in degenerate models. We provide a theoretical explanation for these empirical results, and prove that latent alignment objectives are incompatible with teacher forcing.

📄 PDF Abstract BibTeX arXiv:2104.09554

Code (0)

등록된 구현이 없습니다.

Tasks

Machine TranslationTranslation

Similar Papers 제목 키워드 기반

Non-Monotonic Latent Alignments for CTC-Based Non-Autoregressive Machine Translation

2022-10-08 · Chenze Shao, Yang Feng

Non-autoregressive translation (NAT) models are typically trained with the cross-entropy loss, which forces the model outputs to be aligned verbatim with the target sentence and will highly penalize small shifts in word …

Machine TranslationSentenceTranslation

Non-Autoregressive Machine Translation with Latent Alignments

2020-04-16 · EMNLP 2020 11 · Chitwan Saharia, William Chan, Saurabh Saxena, Mohammad Norouzi

This paper presents two strong methods, CTC and Imputer, for non-autoregressive machine translation that model latent alignments with dynamic programming. We revisit CTC for machine translation and demonstrate that a sim…

Machine TranslationTranslation

CTC-based Non-autoregressive Textless Speech-to-Speech Translation

2024-06-11 · Qingkai Fang, Zhengrui Ma, Yan Zhou, Min Zhang 외

Direct speech-to-speech translation (S2ST) has achieved impressive translation quality, but it often faces the challenge of slow decoding due to the considerable length of speech sequences. Recently, some research has tu…

Knowledge DistillationMachine TranslationSpeech-to-Speech TranslationTranslation

Hint-Based Training for Non-Autoregressive Machine Translation

2019-09-15 · IJCNLP 2019 11 · Zhuohan Li, Zi Lin, Di He, Fei Tian 외

Due to the unparallelizable nature of the autoregressive factorization, AutoRegressive Translation (ART) models have to generate tokens sequentially during decoding and thus suffer from high inference latency. Non-AutoRe…

de-enMachine TranslationTranslation

Hint-based Training for Non-Autoregressive Translation

2019-05-01 · ICLR 2019 5 · Zhuohan Li, Di He, Fei Tian, Tao Qin 외

Machine translation is an important real-world application, and neural network-based AutoRegressive Translation (ART) models have achieved very promising accuracy. Due to the unparallelizable nature of the autoregressive…

de-enMachine TranslationTranslation