paper-with-me

홈 › Papers

Revisiting Checkpoint Averaging for Neural Machine Translation

2022-10-21 · Yingbo Gao, Christian Herold, Zijian Yang, Hermann Ney

Checkpoint averaging is a simple and effective method to boost the performance of converged neural machine translation models. The calculation is cheap to perform and the fact that the translation improvement almost comes for free, makes it widely adopted in neural machine translation research. Despite the popularity, the method itself simply takes the mean of the model parameters from several checkpoints, the selection of which is mostly based on empirical recipes without many justifications. In this work, we revisit the concept of checkpoint averaging and consider several extensions. Specifically, we experiment with ideas such as using different checkpoint selection strategies, calculating weighted average instead of simple mean, making use of gradient information and fine-tuning the interpolation weights on development data. Our results confirm the necessity of applying checkpoint averaging for optimal performance, but also suggest that the landscape between the converged checkpoints is rather flat and not much further improvement compared to simple averaging is to be obtained.

📄 PDF Abstract BibTeX arXiv:2210.11803

Code (0)

등록된 구현이 없습니다.

Tasks

Machine TranslationTranslation

Similar Papers 제목 키워드 기반

The CUED's Grammatical Error Correction Systems for BEA-2019

2019-06-29 · WS 2019 8 · Felix Stahlberg, Bill Byrne

We describe two entries from the Cambridge University Engineering Department to the BEA 2019 Shared Task on grammatical error correction. Our submission to the low-resource track is based on prior work on using finite st…

Grammatical Error CorrectionMachine TranslationTranslation

Training Tips for the Transformer Model

2018-04-01 · Martin Popel, Ondřej Bojar

This article describes our experiments in neural machine translation using the recent Tensor2Tensor framework and the Transformer sequence-to-sequence model (Vaswani et al., 2017). We examine some of the critical paramet…

Machine TranslationmodelSentenceTranslation

Fine-tuning on Clean Data for End-to-End Speech Translation: FBK @ IWSLT 2018

2018-10-16 · IWSLT (EMNLP) 2018 10 · Mattia Antonino Di Gangi, Roberto Dessì, Roldano Cattoni, Matteo Negri 외

This paper describes FBK's submission to the end-to-end English-German speech translation task at IWSLT 2018. Our system relies on a state-of-the-art model based on LSTMs and CNNs, where the CNNs are used to reduce the t…

Machine TranslationTranslation

LiLiuM: eBay's Large Language Models for e-commerce

2024-06-17 · Christian Herold, Michael Kozielski, Leonid Ekimov, Pavel Petrushkov 외

We introduce the LiLiuM series of large language models (LLMs): 1B, 7B, and 13B parameter models developed 100% in-house to fit eBay's specific needs in the e-commerce domain. This gives eBay full control over all aspect…

Language ModelingLanguage ModellingMachine TranslationNatural Language Understanding+2

Growing Together: Modeling Human Language Learning With n-Best Multi-Checkpoint Machine Translation

2020-06-07 · WS 2020 7 · El Moatez Billah Nagoudi, Muhammad Abdul-Mageed, Hasan Cavusoglu

We describe our submission to the 2020 Duolingo Shared Task on Simultaneous Translation And Paraphrase for Language Education (STAPLE) (Mayhew et al., 2020). We view MT models at various training stages (i.e., checkpoint…

Machine TranslationTranslation