Learning Associative Inference Using Fast Weight Memory

Benchmark Model Rank Results
language-modelling-on-penn-treebank-wordAWD-FWM Schlag et al. (2020)#20Test perplexity: 54.48Validation perplexity: 56.76Params: 24M
language-modelling-on-wikitext-2AWD-FWM Schlag et al. (2020)#26Test perplexity: 61.65Validation perplexity: 54.48