XLNet-Large (ensemble)

Reported on 5 benchmarks across 3 tasks · 1 paper · 4 SOTA

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

Natural Language Processing5 results

Semantic Textual SimilarityonQuora Question Pairs
Accuracy· 2019-06-19
90.3
best: 92.4 (data2vec)
SOTA
XLNet: Generalized Autoregressive Pretraining for Language Understanding arXiv:1906.08237
Semantic Textual SimilarityonQuora Question Pairs
F1· 2019-06-19
74.2
best: 90.7 (ALICE)
SOTA
XLNet: Generalized Autoregressive Pretraining for Language Understanding arXiv:1906.08237
Paraphrase IdentificationonQuora Question Pairs
Accuracy· 2019-06-19
90.3
best: 92.4 (data2vec)
SOTA
XLNet: Generalized Autoregressive Pretraining for Language Understanding arXiv:1906.08237
Paraphrase IdentificationonQuora Question Pairs
F1· 2019-06-19
74.2
best: 90.7 (ALICE)
SOTA
XLNet: Generalized Autoregressive Pretraining for Language Understanding arXiv:1906.08237
Sentiment AnalysisonSST-2 Binary classification
Accuracy· 2019-06-19
96.8
best: 97.5 (T5-11B)
XLNet: Generalized Autoregressive Pretraining for Language Understanding arXiv:1906.08237