conftrace_

sequence loss

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

noise-robust training (1) reducible holdout loss (1) noise-robust learning (1) edit-invariant loss (1) non-autoregressive generation (1) language model pretraining (1) joint training (1) slot filling (1) intent recognition (1) pre-trained language model (1)

Papers

Sequence Reducible Holdout Loss for Language Model Pretraining COLING 2024 Don’t Take It Literally: An Edit-Invariant Sequence Loss for Text Generation NAACL 2022 On joint training with interfaces for spoken language understanding Interspeech 2022