|
Training Transformer on small dataset of source code sequences
|
|
1
|
849
|
September 8, 2020
|
|
SentencePiece: Understanding pre-tokenization on raw Unicode bytes?
|
|
2
|
1067
|
August 26, 2020
|
|
Language Model Accuracy
|
|
1
|
728
|
August 21, 2020
|
|
OpenNMT with CTranslate2: the most efficient NMT system at WNGT 2020
|
|
1
|
1547
|
July 28, 2020
|
|
Should you scale step-based parameters with the number of devices you use?
|
|
2
|
676
|
July 27, 2020
|
|
Translate from multiple paragraphs to sentence
|
|
2
|
471
|
July 16, 2020
|
|
Multilingual training experiments
|
|
7
|
2227
|
June 25, 2020
|
|
Doubt with overview of NMT Pipeline
|
|
4
|
914
|
June 16, 2020
|
|
Cleaning and Normalizing WMT Dataset for French to English
|
|
0
|
784
|
June 16, 2020
|
|
BPE vs Word tokenization for Preprocessing
|
|
1
|
1584
|
June 15, 2020
|
|
How can I get some hidden states of model(such as self-attention matrix) during translation
|
|
3
|
765
|
June 10, 2020
|
|
Are you interested in training Russian-Abkhazian parallel corpus?
|
|
29
|
2721
|
June 9, 2020
|
|
Length Penalty Question
|
|
5
|
2727
|
June 2, 2020
|
|
Model "jamming" on words
|
|
2
|
550
|
May 18, 2020
|
|
Sentence meaning score
|
|
1
|
520
|
May 18, 2020
|
|
Effectiveness of Transformer model on small dataset
|
|
7
|
4507
|
May 12, 2020
|
|
Doubt with Number of Epoch
|
|
1
|
597
|
May 11, 2020
|
|
Transformers on low resource corpora
|
|
5
|
1709
|
May 4, 2020
|
|
Bad data in WMT14 en-de
|
|
1
|
489
|
April 23, 2020
|
|
Are composite tokens possible?
|
|
1
|
558
|
April 17, 2020
|
|
Which BLEU script to use?
|
|
1
|
1957
|
April 17, 2020
|
|
What's the best en-de WMT14 BLEU in onmt?
|
|
1
|
491
|
April 9, 2020
|
|
Multifeature translation question
|
|
18
|
1899
|
March 30, 2020
|
|
Quality/Confidence score
|
|
0
|
633
|
March 11, 2020
|
|
Qs about NMT learning
|
|
7
|
573
|
March 6, 2020
|
|
Looking for a master's thesis topic
|
|
6
|
1870
|
March 4, 2020
|
|
In a Transformer model, why does one sum positional encoding to the embedding rather than concatenate it?
|
|
2
|
2208
|
February 17, 2020
|
|
CCMatrix: A billion-scale bitext data set for training translation models
|
|
2
|
1074
|
February 12, 2020
|
|
Trainings steps Q
|
|
1
|
471
|
February 10, 2020
|
|
Types of generalizations learned by NMT?
|
|
0
|
508
|
February 10, 2020
|