Understanding and Improving Lexical Choice in Non-Autoregressive Translation
May 3, 2021
Speakers
Liang Ding
Řečník · 1 sledující
Longyue Wang
Řečník · 1 sledující
Derek F. Wong
Řečník · 0 sledujících
Dacheng Tao
Řečník · 1 sledující
Zhaopeng Tu
Řečník · 2 sledující
About
Knowledge distillation (KD) is essential for training non-autoregressive translation (NAT) models by reducing the complexity of the raw data with an autoregressive teacher model. In this study, we empirically show that as a side effect of this training, the lexical choice errors on low-frequency words are propagated to the NAT model from the teacher model. To alleviate this problem, we propose to expose the raw data to NAT models to restore the useful information of low-frequency words, which are missed in the distilled data. To this end, we introduce an extra Kullback-Leibler divergence term derived by comparing the lexical choice of NAT model and that embedded in the raw data. Experimental results across language pairs and model architectures demonstrate the effectiveness and universality of the proposed approach. Extensive analyses confirm our claim that our approach improves performance by reducing the lexical choice errors on low-frequency words. Encouragingly, our approach pushes the SOTA NAT performance on the WMT14 English-German and WMT16 Romanian-English datasets up to 27.8 and 33.8 BLEU points, respectively. The codes and models will be released.Knowledge distillation (KD) is essential for training non-autoregressive translation (NAT) models by reducing the complexity of the raw data with an autoregressive teacher model. In this study, we empirically show that as a side effect of this training, the lexical choice errors on low-frequency words are propagated to the NAT model from the teacher model. To alleviate this problem, we propose to expose the raw data to NAT models to restore the useful information of low-frequency words, which ar…
About ICLR 2021
The International Conference on Learning Representations (ICLR) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence called representation learning, but generally referred to as deep learning. ICLR is globally renowned for presenting and publishing cutting-edge research on all aspects of deep learning used in the fields of artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, text understanding, gaming, and robotics.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Dynamic Tensor Rematerialization
Zhlédnout později
Learning Energy-Based Models by Diffusion Recovery Likelihood
Zhlédnout později
Ruiqi Gao, …
Few-Shot Bayesian Optimization with Deep Kernel Surrogates
Zhlédnout později
Robust Conversational AI with Grounded Text Generation
Zhlédnout později
Jianfeng Gao, …
Co-clustering of evolving count matrices in pharmacovigilance with the dynamic latent block model
Zhlédnout později
