AWD-LSTM-DOC

Reported on 4 benchmarks across 1 task · 1 paper

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· 2018-08-30
52.38
best: 20.5 (GPT-3 (Zero-Shot))
Direct Output Connection for a High-Rank Language Model arXiv:1808.10143
Language ModellingonPenn Treebank (Word Level)
Validation perplexity· 2018-08-30
54.12
best: 36.1 (BERT-Large-CAS)
Direct Output Connection for a High-Rank Language Model arXiv:1808.10143
Language ModellingonWikiText-2
Test perplexity· 2018-08-30
58.03
best: 8.21 (SparseGPT (175B, 50% Sparsity))
Direct Output Connection for a High-Rank Language Model arXiv:1808.10143
Language ModellingonWikiText-2
Validation perplexity· 2018-08-30
60.29
best: 15.69 (GPT-2 (fine-tuned))
Direct Output Connection for a High-Rank Language Model arXiv:1808.10143