1 |
Quality Assurance of Generative Dialog Models in an Evolving Conversational Agent Used for Swedish Language Practice ...
|
|
|
|
BASE
|
|
Show details
|
|
2 |
Slangvolution: A Causal Analysis of Semantic Change and Frequency Dynamics in Slang ...
|
|
|
|
BASE
|
|
Show details
|
|
3 |
Similarity between person roles in a card sorting experiment ...
|
|
|
|
BASE
|
|
Show details
|
|
4 |
SPT-Code: Sequence-to-Sequence Pre-Training for Learning Source Code Representations ...
|
|
|
|
BASE
|
|
Show details
|
|
6 |
Ensemble of Opinion Dynamics Models to Understand the Role of the Undecided in the Vaccination Debate ...
|
|
|
|
BASE
|
|
Show details
|
|
9 |
Generating Authentic Adversarial Examples beyond Meaning-preserving with Doubly Round-trip Translation ...
|
|
|
|
BASE
|
|
Show details
|
|
10 |
Pirá: A Bilingual Portuguese-English Dataset for Question-Answering about the Ocean ...
|
|
|
|
BASE
|
|
Show details
|
|
11 |
A comparative study of several parameterizations for speaker recognition ...
|
|
|
|
BASE
|
|
Show details
|
|
12 |
A Neural Pairwise Ranking Model for Readability Assessment ...
|
|
|
|
BASE
|
|
Show details
|
|
13 |
A bilingual approach to specialised adjectives through word embeddings in the karstology domain ...
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Speaker verification in mismatch training and testing conditions ...
|
|
|
|
BASE
|
|
Show details
|
|
15 |
Universal Conditional Masked Language Pre-training for Neural Machine Translation ...
|
|
|
|
Abstract:
Pre-trained sequence-to-sequence models have significantly improved Neural Machine Translation (NMT). Different from prior works where pre-trained models usually adopt an unidirectional decoder, this paper demonstrates that pre-training a sequence-to-sequence model but with a bidirectional decoder can produce notable performance gains for both Autoregressive and Non-autoregressive NMT. Specifically, we propose CeMAT, a conditional masked language model pre-trained on large-scale bilingual and monolingual corpora in many languages. We also introduce two simple but effective methods to enhance the CeMAT, aligned code-switching & masking and dynamic dual-masking. We conduct extensive experiments and show that our CeMAT can achieve significant performance improvement for all scenarios from low- to extremely high-resource languages, i.e., up to +14.4 BLEU on low resource and +7.9 BLEU improvements on average for Autoregressive NMT. For Non-autoregressive NMT, we demonstrate it can also produce consistent ... : Accepted to ACL 2022 Main conference ...
|
|
Keyword:
Computation and Language cs.CL; FOS Computer and information sciences
|
|
URL: https://dx.doi.org/10.48550/arxiv.2203.09210 https://arxiv.org/abs/2203.09210
|
|
BASE
|
|
Hide details
|
|
16 |
SMDT: Selective Memory-Augmented Neural Document Translation ...
|
|
|
|
BASE
|
|
Show details
|
|
17 |
Learning How to Translate North Korean through South Korean ...
|
|
|
|
BASE
|
|
Show details
|
|
18 |
When do Contrastive Word Alignments Improve Many-to-many Neural Machine Translation? ...
|
|
|
|
BASE
|
|
Show details
|
|
19 |
Conditional Bilingual Mutual Information Based Adaptive Training for Neural Machine Translation ...
|
|
|
|
BASE
|
|
Show details
|
|
|
|