Aligned Cross Entropy for Non-Autoregressive Machine Translation

被引:0
|
作者
Ghazvininejad, Marjan [1 ]
Karpukhin, Vladimir [1 ]
Zettlemoyer, Luke [1 ]
Levy, Omer [1 ]
机构
[1] Facebook AI Res, Menlo Pk, CA 94025 USA
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Non-autoregressive machine translation models significantly speed up decoding by allowing for parallel prediction of the entire target sequence. However, modeling word order is more challenging due to the lack of autoregressive factors in the model. This difficultly is compounded during training with cross entropy loss, which can highly penalize small shifts in word order. In this paper, we propose aligned cross entropy (AXE) as an alternative loss function for training of non-autoregressive models. AXE uses a differentiable dynamic program to assign loss based on the best possible monotonic alignment between target tokens and model predictions. AXE-based training of conditional masked language models (CMLMs) substantially improves performance on major WMT benchmarks, while setting a new state of the art for non-autoregressive models.
引用
收藏
页数:9
相关论文
共 50 条
  • [31] Correcting translation for non-autoregressive transformer
    Wang, Shuheng
    Huang, Heyan
    Shi, Shumin
    Li, Dongbai
    Guo, Dongen
    APPLIED SOFT COMPUTING, 2025, 168
  • [32] A Study of Syntactic Multi-Modality in Non-Autoregressive Machine Translation
    Zhang, Kexun
    Wang, Rui
    Tan, Xu
    Guo, Junliang
    Ren, Yi
    Qin, Tao
    Liu, Tie-Yan
    NAACL 2022: THE 2022 CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: HUMAN LANGUAGE TECHNOLOGIES, 2022, : 1747 - 1757
  • [33] Alleviating repetitive tokens in non-autoregressive machine translation with unlikelihood training
    Wang, Shuheng
    Shi, Shumin
    Huang, Heyan
    SOFT COMPUTING, 2024, 28 (5) : 4681 - 4688
  • [34] Alleviating repetitive tokens in non-autoregressive machine translation with unlikelihood training
    Shuheng Wang
    Shumin Shi
    Heyan Huang
    Soft Computing, 2024, 28 : 4681 - 4688
  • [35] Revisiting Non-Autoregressive Translation at Scale
    Wang, Zhihao
    Wang, Longyue
    Su, Jinsong
    Yao, Junfeng
    Tu, Zhaopeng
    FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2023), 2023, : 12051 - 12065
  • [36] Sequence-Level Training for Non-Autoregressive Neural Machine Translation
    Shao, Chenze
    Feng, Yang
    Zhang, Jinchao
    Meng, Fandong
    Zhou, Jie
    COMPUTATIONAL LINGUISTICS, 2021, 47 (04) : 891 - 925
  • [37] Improving Non-autoregressive Machine Translation with Error Exposure and Consistency Regularization
    Chen, Xinran
    Duan, Sufeng
    Liu, Gongshen
    NATURAL LANGUAGE PROCESSING AND CHINESE COMPUTING, PT III, NLPCC 2024, 2025, 15361 : 240 - 252
  • [38] Iterative Refinement in the Continuous Space for Non-Autoregressive Neural Machine Translation
    Lee, Jason
    Shu, Raphael
    Cho, Kyunghyun
    PROCEEDINGS OF THE 2020 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING (EMNLP), 2020, : 1006 - 1015
  • [39] Non-autoregressive Machine Translation with Probabilistic Context-free Grammar
    Gui, Shangtong
    Shao, Chenze
    Ma, Zhengrui
    Zhang, Xishan
    Chen, Yunji
    Feng, Yang
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [40] Guiding Non-Autoregressive Neural Machine Translation Decoding with Reordering Information
    Ran, Qiu
    Lin, Yankai
    Li, Peng
    Zhou, Jie
    THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2021, 35 : 13727 - 13735