Feedback Transformer

Reported on 2 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.

Medical2 results

Language ModellingonPenn Treebank (Character Level)
Bit per Character (BPC)· 2020-02-21
1.16
best: 1.38 (Bipartite Flow)
Addressing Some Limitations of Transformers with Feedback Memory arXiv:2002.09402
Language Modellingonenwik8
Bit per Character (BPC)· 2020-02-21
0.96
best: 1.67 (LSTM (7 layers))
Addressing Some Limitations of Transformers with Feedback Memory arXiv:2002.09402