AWD-LSTM + continuous cache pointer

Reported on 4 benchmarks across 1 task · 1 paper · 4 SOTA

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

Medical4 results

Language ModellingonPenn Treebank (Word Level)
Test perplexity· 2017-08-07
52.8
best: 20.5 (GPT-3 (Zero-Shot))
SOTA
Regularizing and Optimizing LSTM Language Models arXiv:1708.02182
Language ModellingonPenn Treebank (Word Level)
Validation perplexity· 2017-08-07
53.9
best: 36.1 (BERT-Large-CAS)
SOTA
Regularizing and Optimizing LSTM Language Models arXiv:1708.02182
Language ModellingonWikiText-2
Test perplexity· 2017-08-07
52
best: 8.21 (SparseGPT (175B, 50% Sparsity))
SOTA
Regularizing and Optimizing LSTM Language Models arXiv:1708.02182
Language ModellingonWikiText-2
Validation perplexity· 2017-08-07
53.8
best: 15.69 (GPT-2 (fine-tuned))
SOTA
Regularizing and Optimizing LSTM Language Models arXiv:1708.02182