Sign in

From Distillation to Hard Negative Sampling: Making Sparse Neural IR Models More Effective.

Thibault FormalCarlos LassanceBenjamin PiwowarskiStéphane Clinchant
Published in: SIGIR (2022)
Keyphrases
  • information retrieval
  • neural network
  • language model
  • machine learning
  • high dimensional
  • active learning
  • data points
  • vector space