Stacked Bi-LSTMs (shortcut connections, max-pooling, attention)

Reported on 3 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.

Natural Language Processing3 results

Natural Language InferenceonSNLI
% Test Accuracy· 2018-11-02
84.4
best: 94.7 (UnitedSynT5 (3B))
Combining Similarity Features and Deep Representation Learning for Stance Detection in the Context of Checking Fake News arXiv:1811.00706
Natural Language InferenceonMultiNLI
Matched· 2018-11-02
70.7
best: 92.6 (Turing NLR v5 XXL 5.4B (fine-tuned))
Combining Similarity Features and Deep Representation Learning for Stance Detection in the Context of Checking Fake News arXiv:1811.00706
Natural Language InferenceonMultiNLI
Mismatched· 2018-11-02
70.5
best: 92.4 (Turing NLR v5 XXL 5.4B (fine-tuned))
Combining Similarity Features and Deep Representation Learning for Stance Detection in the Context of Checking Fake News arXiv:1811.00706