AWD-FWM Schlag et al. (2020)

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· 2020-11-16
54.48
best: 20.5 (GPT-3 (Zero-Shot))
Learning Associative Inference Using Fast Weight Memory arXiv:2011.07831
Language ModellingonPenn Treebank (Word Level)
Validation perplexity· 2020-11-16
56.76
best: 36.1 (BERT-Large-CAS)
Learning Associative Inference Using Fast Weight Memory arXiv:2011.07831
Language ModellingonWikiText-2
Test perplexity· 2020-11-16
61.65
best: 8.21 (SparseGPT (175B, 50% Sparsity))
Learning Associative Inference Using Fast Weight Memory arXiv:2011.07831
Language ModellingonWikiText-2
Validation perplexity· 2020-11-16
54.48
best: 15.69 (GPT-2 (fine-tuned))
Learning Associative Inference Using Fast Weight Memory arXiv:2011.07831