FNet-Large

Reported on 6 benchmarks across 4 tasks · 1 paper

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

Natural Language Processing6 results

Natural Language InferenceonMultiNLI
Matched· 2021-05-09
78
best: 92.6 (Turing NLR v5 XXL 5.4B (fine-tuned))
FNet: Mixing Tokens with Fourier Transforms arXiv:2105.03824
Natural Language InferenceonMultiNLI
Mismatched· 2021-05-09
76
best: 92.4 (Turing NLR v5 XXL 5.4B (fine-tuned))
FNet: Mixing Tokens with Fourier Transforms arXiv:2105.03824
Semantic Textual SimilarityonSTS Benchmark
Spearman Correlation· 2021-05-09
0.84
best: 0.931 (Mnet-Sim)
FNet: Mixing Tokens with Fourier Transforms arXiv:2105.03824
Semantic Textual SimilarityonQuora Question Pairs
F1· 2021-05-09
85
best: 90.7 (ALICE)
FNet: Mixing Tokens with Fourier Transforms arXiv:2105.03824
Sentiment AnalysisonSST-2 Binary classification
Accuracy· 2021-05-09
94
best: 97.5 (T5-11B)
FNet: Mixing Tokens with Fourier Transforms arXiv:2105.03824
Paraphrase IdentificationonQuora Question Pairs
F1· 2021-05-09
85
best: 90.7 (ALICE)
FNet: Mixing Tokens with Fourier Transforms arXiv:2105.03824