zomux lanmt-ebm issues - Githubissues

zomux / lanmt-ebm

lanmt ebm

11 stars 1 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Parallel decoding commands

#27 zomux opened 4 years ago
2
Train EBM on WMT'14 En-De

#26 jaseleephd opened 4 years ago
1
Retrain NAT model on IWSLT16

#25 zomux opened 4 years ago
1
Magnitude is not converging

#24 zomux opened 4 years ago
0
Proper comparison between energy vs. score parameterization on IWSLT

#23 jaseleephd opened 4 years ago
0
Proper speed analysis on IWSLT

#22 jaseleephd opened 4 years ago
1
WMT'16 En->Ro

#21 jaseleephd opened 4 years ago
11
Paper writing

#20 zomux opened 4 years ago
1
Score matching loss for denoising models

#19 zomux opened 4 years ago
5
Test all energy model settings on WMT14 dataset

#18 zomux opened 4 years ago
4
Better base latent model

#17 zomux opened 4 years ago
14
Learn from corruption loss in VAE

#16 zomux opened 4 years ago
0
Joint training of auto-encoder and EBM

#15 zomux opened 4 years ago
0
Snippet: ABCI job launcher

#14 zomux opened 4 years ago
0
General Issues

#13 zomux opened 4 years ago
0
Independent Energy-based MT

#12 zomux opened 4 years ago
1
Direct cross-entropy loss (applied on language model)

#11 zomux opened 4 years ago
10
Discrete EBM language model

#10 zomux opened 4 years ago
2
Denoising model results

#9 zomux opened 4 years ago
0
A summarization of all losses

#8 zomux opened 4 years ago
5
EBM language model training

#7 zomux opened 4 years ago
3
universal target-side latent model and EBM training

#6 zomux opened 4 years ago
4
fix the bug in TransformerCrossEncoderLayer

#5 zomux opened 4 years ago
1
Taylor gradient matching loss

#4 zomux opened 4 years ago
0
BLEU approximating EBM

#3 zomux opened 4 years ago
3
retrain q(z|x,y) and p(z|x) with convolution/disentangled models

#2 zomux opened 4 years ago
3
Convolution-based energy model (delta inference approximation)

#1 zomux opened 4 years ago
7