Inan et al. (2016) - Variational LSTM (tied) (h=650) + augmented loss

Reported on 2 benchmarks across 1 task · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Medical2 results

Language ModellingonWikiText-2
Test perplexity· 2016-11-04
87
best: 8.21 (SparseGPT (175B, 50% Sparsity))
SOTA
Tying Word Vectors and Word Classifiers: A Loss Framework for Language Modeling arXiv:1611.01462
Language ModellingonWikiText-2
Validation perplexity· 2016-11-04
91.5
best: 15.69 (GPT-2 (fine-tuned))
SOTA
Tying Word Vectors and Word Classifiers: A Loss Framework for Language Modeling arXiv:1611.01462