T5-Large 770M

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

Question AnsweringonSQuAD1.1 dev
EM· 2019-10-23
86.66
best: 90.06 (T5-11B)
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer arXiv:1910.10683
Question AnsweringonSQuAD1.1 dev
F1· 2019-10-23
93.79
best: 95.77 (XLNet+DSC)
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer arXiv:1910.10683
Natural Language InferenceonWNLI
Accuracy· 2019-10-23
85.6
best: 95.9 (Turing NLR v5 XXL 5.4B (fine-tuned))
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer arXiv:1910.10683
Natural Language InferenceonMultiNLI
Mismatched· 2019-10-23
89.6
best: 92.4 (Turing NLR v5 XXL 5.4B (fine-tuned))
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer arXiv:1910.10683
Semantic Textual SimilarityonSTS Benchmark
Spearman Correlation· 2019-10-23
0.886
best: 0.931 (Mnet-Sim)
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer arXiv:1910.10683
Sentiment AnalysisonSST-2 Binary classification
Accuracy· 2019-10-23
96.3
best: 97.5 (T5-11B)
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer arXiv:1910.10683