Injecting WordNet lexical relations as bias terms into multi-head attention improves accuracy on the adversarial SNLI test set, with BERT reaching 94.1%, equal to estimated human performance.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Knowledge Enhanced Attention for Robust Natural Language Inference
Injecting WordNet lexical relations as bias terms into multi-head attention improves accuracy on the adversarial SNLI test set, with BERT reaching 94.1%, equal to estimated human performance.