Login / Signup

Outlier Reduction with Gated Attention for Improved Post-training Quantization in Large Sequence-to-sequence Speech Foundation Models.

Dominik WagnerIlja BaumannKorbinian RiedhammerTobias Bocklet
Published in: CoRR (2024)
Keyphrases
  • statistical models
  • artificial neural networks
  • prior knowledge
  • probabilistic model
  • machine learning
  • training data
  • training set
  • supervised learning
  • statistical model
  • computational models
  • training process