Login / Signup

Attention Weights in Transformer NMT Fail Aligning Words Between Sequences but Largely Explain Model Predictions.

Javier FerrandoMarta R. Costa-jussà
Published in: EMNLP (Findings) (2021)
Keyphrases
  • computational model
  • weighting scheme
  • cost function
  • theoretical framework
  • machine learning
  • objective function
  • expert systems
  • probabilistic model